Office 365 for IT Pros Companion

Volume 2019 Tony Redmond

Visit to download the full and correct content document:
https://textbookfull.com/product/office-365-for-it-pros-companion-volume-2019-tony-re
dmond/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Office 365 for IT Pros Fifth Edition Tony Redmond

https://textbookfull.com/product/office-365-for-it-pros-fifth-
edition-tony-redmond/

Mastering VBA 2019 For Microsoft Office 365 2019

Edition Richard Mansfield

https://textbookfull.com/product/mastering-vba-2019-for-
microsoft-office-365-2019-edition-richard-mansfield/

Illustrated Microsoft Office 365 Office 2019

Intermediate 1st Edition Jennifer Duffy

https://textbookfull.com/product/illustrated-microsoft-
office-365-office-2019-intermediate-1st-edition-jennifer-duffy/

Microsoft Office 365 for dummies Reed

https://textbookfull.com/product/microsoft-office-365-for-
dummies-reed/
Essential PowerShell for Office 365 Vlad Catrinescu

https://textbookfull.com/product/essential-powershell-for-
office-365-vlad-catrinescu/

Office 365 for Healthcare Professionals Nidhish Dhru

https://textbookfull.com/product/office-365-for-healthcare-
professionals-nidhish-dhru/

MCA Microsoft Office Specialist Office 365 and Office

2019 Study Guide Excel Associate Exam MO 200 1st
Edition Butow

https://textbookfull.com/product/mca-microsoft-office-specialist-
office-365-and-office-2019-study-guide-excel-associate-exam-
mo-200-1st-edition-butow/

Microsoft Excel Functions and Formulas with Excel 2019

Office 365 5th Edition Bernd Held

https://textbookfull.com/product/microsoft-excel-functions-and-
formulas-with-excel-2019-office-365-5th-edition-bernd-held/

From IT Pro to Cloud Pro Microsoft Office 365 and

SharePoint Online 1st Edition Ben Curry

https://textbookfull.com/product/from-it-pro-to-cloud-pro-
microsoft-office-365-and-sharepoint-online-1st-edition-ben-curry/
Office 365 for IT Professionals (2019 Edition)
Published by Tony Redmond
© Copyright 2015-2018 by Tony Redmond, Michael Van Horenbeeck, and Paul
Cunningham.
All rights reserved. No part of this book may be reproduced or transmitted in any
form or by any means without the written permission of the authors.
The example companies, organizations, products, domain names, email
addresses, logos, people, places and event depicted herein are fictitious. No
association with any real company, organization, people, domain name, email
address, logo, person, place, or event is intended or should be inferred. The book
expresses the views and opinions of the authors. The information presented in
the book is provided without any express, statutory, or implied warranties. The
authors cannot be held liable for any damages caused or alleged to be caused
either directly or indirectly by this book.
Although the authors are members of Microsoft’s Most Valuable Professional
(MVP) program, the content of this book solely represents their views and
opinions about Office 365 and any other technologies mentioned in the text and
is not endorsed in any way by Microsoft Corporation.
Please be respectful of the rights of the authors and do not make copies of
this eBook available to others.
Fifth (2019) edition. Previous editions:

Office 365 for Exchange Professionals (May 2015 and September 2015).
Office 365 for IT Pros (3rd edition – June 2016).]
Office 365 for IT Pros (4th edition – June 2017).

This is the companion volume for Office 365 for IT Pros (2019 edition). Its
content is valuable, but we do not update it as often as we do for material in the
main book.
This is update 1 published on 1 July 2018. You can find information about the
changes made in each update in our change log.
Chapter 1: Introduction
Tony Redmond

Welcome to the Companion Volume

The fourth edition of Office 365 for IT Pros reached 1,150 pages. Some of the
material had aged a little, and some of it was not as interesting to everyone. So,
we took the decision to create this companion volume and use it as a home for
information that we think is still valuable and should be shared, but might not
warrant a place in the main book.

Office 365 History

Office 365 is not Microsoft’s first cloud platform. In fact, Microsoft got into the
cloud application game in 2005 when it started to provide a managed service for
Exchange to some customers. The first public information about this effort
appeared in October 2007 when Microsoft announced “Exchange Labs”. Among
other features, Microsoft said that Exchange Labs supported 5 GB mailboxes
and used SSL to secure client communications. The commercial launch followed
when Microsoft brought this effort forward into Business Productivity Online
Services (BPOS), launched in 2008. BPOS included Exchange Online,
SharePoint Online, Office Communications Server Online (an ancestor of Skype
for Business), Forefront (anti-virus), and Office Live Meeting. Collectively, this
set of applications was referred to as Microsoft Online Services.
The technology used by BPOS was an adapted form of the software sold at the
time to on-premises customers as Exchange 2007, SharePoint 2007, and so on.
The big difference between Office 365 and BPOS is the stability and robustness
of the current platform, largely due to the maturity of the workloads now running
inside Office 365 combined with a highly developed automation framework to
orchestrate operations. In a nutshell, BPOS was an example of how to take
software designed to be deployed in a traditional on-premises environment and
bend it so that it functioned in a multi-tenant infrastructure accessed through the
Internet. The problem was that “bending” the software often made things break
simply because the software was not designed to cope with the stresses and
pressures generated by large-scale multi-tenant operations.
Office 365 is built around software designed with the unique demand of cloud-
scale operations in mind, so its delivery and reliability record is much better than
BPOS ever managed. Still, BPOS proved to be extraordinary useful to Microsoft
in educating engineers about how to build software to function in cloud
environments, even if its reputation suffered due to the relatively poor
performance against the Service Level Agreements (SLA) negotiated with
customers. The value of attributes such as automation, simplification, and
standardization quickly became recognized and valued through working through
the trials and tribulations of BPOS. What’s absolutely certain is that without the
experience gained through working through real-life customer deployments of
BPOS, Microsoft could not have achieved as much as they have with Office 365
since.

A remnant of Exchange Labs: If you look at the properties of a mailbox, you’ll

see that the LegacyExchangeDN property still has its roots in Exchange Labs.
For example, you might see a value like this:

/o=ExchangeLabs/ou=Exchange Administrative Group (FYDIBOHF23SPDLT)/

cn=Recipients/cn=yourdomain.onmicrosoft.com-52094eea20

Apart from Exchange Labs, the other bit of Exchange history is the fact that an
administrative group is still specified. Administrative groups appeared as a unit
of server management in Exchange 2000 and were phased out in Exchange
2007. However, the LegacyExchangeDN property goes back even further to the
X.500-like directory structure used by the first generation of Exchange (4.0 to
5.5 in 1996-99). All of this goes to show that you can’t really discard history too
easily.
Although BPOS provided a healthy dose of customer reality to Microsoft’s
engineers, you cannot take too many risks with a production offering. Microsoft
therefore continued to operate “Exchange Labs” alongside BPOS to conduct
experiments at scale and learn just what it took to transform Exchange from a
somewhat stodgy but powerful email and collaboration server into software that
could function in the cloud. Exchange Labs was the playground while BPOS
attempted – sometimes quite well – to deliver a commercial offering, even if it
was handicapped by its enterprise heritage. Together, BPOS and Exchange Labs
constituted the greenhouse for what later evolved into Exchange Online and an
operations framework that permeates throughout Office 365.
During the same period as they were figuring out how to execute their initial
cloud deployment strategy, the Exchange product group made a big bet on
PowerShell by using it for many administrative operations in Exchange 2007.
Exchange was the first major Microsoft server application to use PowerShell so
extensively and did so at a time when “Monad” (the original code name) was
often derided as a poor combination of UNIX-style scripting and impenetrable
syntax. As it turned out, PowerShell has made a huge contribution to the success
of Exchange in the cloud and Microsoft uses PowerShell scripts to automate
many common administrative operations that run inside Office 365.

Office 365 is not Office 365: The power of marketing means that the Office
365 name is applied to a number of Microsoft offerings. This book is about
Office 365 for business and not Office 365 Home (a five-user license for the
Office desktop applications) or Office 365 Personal (a one-user license for the
Office desktop applications). Both are worthy in their own right, but they have
nothing to do with Office 365 cloud-based application services like Exchange
Online.

Technical steps along the path to the cloud

Important as PowerShell is to Exchange Online, it’s not the only area of
technical innovation that has allowed Microsoft to transform the enterprise-
centric traditional-deployment model used by Microsoft on-premises server
products into today's Office 365. Many threads have come together to deliver
that work, many of which started in the 2003-2005 timeframe. Although you
might delight in the notion of a grand technical plan that has come together to
deliver Office 365, it’s much more of a case of blessed serendipity allied to some
great engineering and visionary work delivered over a series of server versions.
The two base Office 365 workloads are Exchange and SharePoint. Thw two
products share a common history in that SharePoint Portal Server 2001, the first
version of SharePoint, used the Exchange ESE database engine. SharePoint
subsequently moved to the SQL database engine and has used it since. Each of
the two workloads has its own history and timeline in progressing from a
product embedded in an on-premises ecosystem to becoming a cornerstone of
Office 365. Table 1-1 outlines the areas of technical innovation in different
versions of Exchange and notes why each area is critical to what we use today in
Office 365.
Technical When
Why important to Office 365
innovation introduced
 The basis for automation of many common
Exchange management operations. Remote PowerShell was
PowerShell
2007 introduced in Exchange 2010 and is used with
Exchange Online and other applications.
Exchange Allows clients to connect to Exchange Online without
Autodiscover
2007 knowing details of server names etc.
Although its importance to Office 365 lessened after
the introduction of the Outlook apps for iOS and
Exchange Android, ActiveSync is the de facto standard for
ActiveSync
2003 SP1 mobile connectivity to
Exchange Online.
Role Based
Allows granular access to user and administrator
Access Exchange
functionality. Also used to control the display of UI in
Control 2010
OWA and the Office 365 administration portals.
(RBAC)
The original Exchange Web client goes back to 1997
and has evolved dramatically since to keep pace with
Outlook Web Exchange web developments. Outlook Web App supports
App 5.0 premium access to Exchange Online from Chrome,
IE, Firefox, and Safari browsers and downgraded
access for other browsers.
Exchange
2003 RPC over HTTP (Outlook Anywhere) removed the
RPC over
HTTP Exchange requirement for VPN connectivity to email servers
2013 SP1 across the Internet. Without it, you’d have to create a
MAPI over VPN to connect to Exchange Online. MAPI over
HTTP HTTP is a more modern and effective replacement.

Allows programmatic access to items in the Exchange

Exchange Store without having to resort to the far more
Web Exchange complicated MAPI API. Although the Microsoft
Services 2007 Graph is a more modern API for Exchange along with
(EWS) other Office 365 components, EWS persists and is
used in migration and backup products for Office 365.
ESE is the engine that lies at the heart of the Exchange
 Information Store and has been extended and refined
Extensible for more than 20 years to arrive at the point where
Storage Exchange 100GB-plus mailboxes are usable. Another important
Engine 4.0 aspect is the work done in the 2004-2012 timeframe to
(ESE) drive the storage I/O profile of Exchange from being a
fat slob to a svelte service and so enables the ability to
exploit low-cost storage that delivers massive
mailboxes at a very low price point.
The first HA implementation allowed just two copies
of a database. The introduction of the Database
Availability Group (DAG) in Exchange 2010
expanded this to 16 copies and introduced features
like the lagged copy, single page patching, automatic
failover, and so on. Exchange 2013 continued to
High Exchange
improve matters with database autoreseed and greater
Availability 2007
automation to manage lagged database copy, including
the introduction of the Replay Lag Manager in
Exchange 2013. The HA features allow Exchange
Online to operate with a basic model of four database
copies spread across two datacenters (or more) and
ensure that the 99.9% SLA is met or exceeded.
Some self-healing capabilities were introduced in
Exchange 2010 but Managed Availability took the
idea that servers could monitor their own health and
Managed Exchange take action when required to fix a failed component to
Availability 2013 a new level. Even if on-premises administrators don’t
like its influence over servers very much, Exchange
Online might not be manageable without this degree
of automation.
Multi-tenant environments operate on a fair usage
Workload
Exchange basis. In other words, the workload of a single tenant
management
2010 should not be able to unduly affect others. Workload
(throttling)
management makes this so.
Modern Without a modernized version, on-premises customers
Exchange
public who had invested heavily in the cockroaches of
2013
folders Exchange would never be able to move to the cloud.
Table 1-1: Technical innovation in Exchange that help Exchange Online work
Some technology developed for Office 365 is difficult to move to on-premises
versions. The automatic filtering of inbound email performed by the Focused
Inbox and the Teams, Planner, and Delve Analytics applications are examples of
Office 365 software that will probably never run in a pure on-premises
environment. On the other hand, as demonstrated by the hybrid features in
SharePoint 2016 (hybrid sites, search, and OneDrive sites), it is possible to
provide data taken from on-premises servers and process them in the cloud.
A number of reasons can exist to prevent the transfer of cloud-based software to
on-premises deployments, including:

The technology is complex to deploy and requires substantial effort to

sustain in production. Integrating different pieces of software together so
that they all work as planned is often difficult when software changes all the
time. Traditional IT discipline focuses on structured updates performed in
change windows, something that doesn't work quite so well given the need
to execute updates across many moving parts.
The technology requires a high cost in infrastructure (servers, network,
storage, and automation) to deploy and keep running, which implies a high
cost barrier. Applications that depend on machine learning and artificial
intelligence are often difficult to deploy in an on-premises environment
because of the resources they consume.
The technology needs to be fine-tuned on an ongoing basis to improve its
performance and, in some cases, accuracy. This work usually requires
engineers to be able to make frequent and ongoing software changes.
Less importantly, the technology requires a skill set that might not be
feasible to expect within customer environments.

Remember, Microsoft is able to operate and develop Office 365 by employing

the full resources of the company in addition to a massive financial commitment
to build out the infrastructure. It’s almost inevitable that some components
developed for and implemented first in Office 365 will prove just too complex to
transfer, but the good thing is that Microsoft has built a strong track record of
transfer from cloud to on-premises and nothing indicates that this trend will not
continue.

The importance of technology transfers to on-premises versions: Although on-

premises customers often complain – sometimes bitterly – about the way that
features show up in Exchange Online and not in the latest on-premises update, it
is undeniable that Microsoft has done a good job of transferring technology
developed to help run Exchange Online at scale to on-premises customers. Most
of the work transferred to date has been directed to Exchange 2013; even more
is included in Exchange 2016. Although the nature of cloud services means that
they will always be ahead of on-premises equivalents, the real question is how
quickly will on-premises customers deploy the new versions to take advantage
of the technology transfer? Traditionally, new server versions take several years
before they reach general deployment across the entire customer base.

Wave 14: Office 365 launches

Microsoft put all of the experience gained in BPOS to advantage when it
designed Office 365. When launched, Office 365 used the “Wave 14” set of
Office server products that shipped to customers as Exchange 2010, SharePoint
2010, Lync 2010, and so on. The big difference was that Microsoft had had
several years of operational experience to better understand the demands of the
cloud. More automation was incorporated into the products, the code base was
simplified, the economics were better, and great attention was paid to all aspects
of design, build, and operation.
Another important point in the evolution of Office 365 was the adoption of a
software development method based on the DevOps concept. In effect, this
means that engineering group responsible for the development of Exchange are
the go-to people for problems. In other words, if code fails then it is the engineer
who wrote or maintains the code who has to wake up from blissful slumber to
fix the problem. There’s no doubt that the direct association between code
quality and responsibility for maintenance influenced the way that engineers
created features. It’s obviously important for engineers to take personal pride in
creating the best possible code at all times, but it becomes terribly personal for
an engineer when they are hauled in at 3am to fix an irritating bug.
Some initial glitches occurred in Office 365 that interrupted service to customers
in August and September 2011, but broadly speaking the performance,
reliability, and scalability of the service has proven to be excellent. We will
discuss how Office 365 measures performance against service level agreements
in Chapter 2.
Exchange 2010 introduced a number of important technical advances that have
contributed greatly to the subsequent success of Exchange Online. The Database
Availability Group (DAG) is the most important because it allows Office 365 to
operate a highly available infrastructure for mailboxes. Exchange Online now
spans several thousand DAGs positioned in Office 365 datacenters around the
world; each database has four copies including a lagged copy; and the high
availability features built into the DAG allows work to be transferred quickly
and dependably to another server should a problem arise. It also underpins the
concept of “native data protection”, meaning that Exchange Online does not use
traditional backups to protect data. Instead, a combination of Exchange features
such as user-driven recoverable items, single item retention, and multiple
database copies protect user data so that it can be recovered in the case of
inadvertent loss.
Apart from making sure that mailboxes stay online, the DAG also contributes to
the economics of Office 365 by allowing the use of inexpensive JBOD disks for
the massive amount of storage needed to allow users keep all the information
they want inside massive mailboxes. The economics are such now that it is much
cheaper to allow users as much storage as they want rather than invest in the
time to manage storage. Gigabytes of storage cost a fraction of a penny per
month (amortized over 24 or 36 months) when bought in the quantities
consumed by cloud datacenters, so the fact that someone is using a 100 GB
mailbox that costs a few cents for the storage is nothing when put into context
with their monthly payment. The same is true for other services like OneDrive
for Business or the consumer email services where the providers are happy to
have users consume large amounts of storage in return for the chance to sell
other services. Of course, cloud providers incur massive additional costs other
than storage, but it is interesting to see how storage has become so cheap and
plentiful in such a short time and the influence this has had on data management.
Since its introduction, the DAG has steadily added features to improve its ability
to support low-cost disks. For instance, single page patching arrived in Wave 14
to allow Exchange to detect and patch corrupt pages that appear in both active
and passive database copies. Without single page patching, human
administrators would have to take problematic databases offline and fix them
manually. Traditionally, this would have been done by restoring a backup copy,
but these don’t exist inside Office 365. Database autoreseed is another example.
Introduced in Wave 15, this feature allows administrators to set aside disk space
that Exchange can use to build a new copy of a failed database. The rebuild
happens automatically, which is exactly what you need when the use of low-cost
disks makes it easy to predict that disk and controller outages will be the most
common form of failure inside Office 365 datacenters. And, as it turns out, they
are.
Microsoft also introduced the Mailbox Replication Service (MRS) in Wave 14.
Not quite as exciting or as technically compelling as the DAG, MRS still plays
an enormous role through its ability to move mailboxes from on-premises
servers to Office 365. The transfer is highly automated, batch-driven, happens in
the background, and includes automatic delta synchronization to maintain the
copied mailboxes in a current state until the switchover occurs. If MRS didn’t
exist, it would be very much harder for large companies to transfer mailboxes to
Office 365.

Wave 15: Cloud-Ready by Design

In late 2013, Microsoft launched the next generation of Office 365 based on the
“Wave 15” set of Office servers. By this time the developers had gained an
enormous amount of operational experience from BPOS and the first iteration of
Office 365 and had factored it into the development of Exchange 2013. Wave 15
marked the first time that Office 365 moved ahead of on-premises products in
terms of introducing new code into production. From this point on, new features
appeared in the cloud first and then in an on-premises release.
A common code base was reintroduced to unite the cloud and on-premises
versions of Exchange and a new supportability model was introduced where
Microsoft shipped a cumulative update to Exchange on-premises customers
every quarter that contained bug fixes and new features proven in Office 365.
On-premises customers didn’t get every new feature because some depended on
non-Exchange components (see below) but a continual flow of information from
“the service” (the term used by Microsoft employees to refer to Office 365) is
used by the developers to improve and refine on-premises Exchange.
Managed Availability is a Wave 15 feature that is also a good example of how
Microsoft has transferred technology from the cloud to on-premises Exchange.
Today, Microsoft operates over 200,000 Exchange servers inside its Office 365
datacenters. It would be impossible to have human administrators monitor the
mailbox and other servers at the scale used by Exchange Online and be expected
to detect and take rapid action when problems occur. Indeed, given the need for
humans to sleep and our tendency to lose interest in boring and repetitive
actions, many issues that occur on servers would go unnoticed. Managed
Availability gathers a vast amount of health signals from its probes running on
every Exchange server, decides whether the data indicates a problem, and
responds to any problem that is found, all without human oversight or
intervention. The idea is that Managed Availability should be able to take care of
routine and common issues, leaving the most complex and difficult problems for
humans to solve.
Features such as database autoreseed, namespace rationalization, and the
simplification of the DAG are other examples of how Office 365 has influenced
the evolution of on-premises Exchange. Customers have also gained through the
ten-year development effort to transform the disk and storage requirements of
Exchange from a point where deployments required expensive “enterprise-class”
disks to today where inexpensive JBOD-style storage is the norm. Microsoft had
to change Exchange to be less of an I/O hog to make it feasible to use Exchange
for cloud-based email. After all, if you need expensive disks, you won’t be able
to offer users 50 GB mailboxes at the kind of price points that Office 365
charges today.
The change that occurred in client focus is also worth noting. Whereas Outlook
remains the single most popular client used to connect to Exchange Online, its
Windows-centric development model means that it is slow to adapt and change,
especially in the context of a fast-moving cloud service. The problem is simple:
it takes Microsoft far more time to update the Outlook user interface to introduce
new features and even longer for customers to deploy the new software to user
desktops than it does to make a change to the browser components that drive
Outlook Web App.
Microsoft transformed Outlook Web App in the Wave 15 release. Some of the
updates seemed retrograde at the time because functionality was reduced and
performance was poor, but change was necessary to deliver a user interface that
was capable of working on PCs, tablets, and “candy bar” smartphones. Over
time, missing features have largely been restored alongside a range of new
features and performance has steadily improved. Indeed, the rate of change in
Outlook Web App and the difference that opened up between the version
available to Office 365 users and that provided to on-premises Exchange
customers made it obvious that Outlook Web App is regarded almost as an
experimentation platform for Office 365. In other words, Outlook Web App is
the client that Microsoft can use to introduce new features in a rapid manner,
even when those features are not fully complete. Outlook is the more popular
client, but it lags in the functionality stakes and is likely to always do so until
customers adapt the “Click to Run” variant and accept the fact that desktop user
interfaces are liable to change on almost a monthly basis – at least, when
connected to Office 365.
After the transit to the Wave 15 products was complete, signs of the growing
maturity of the platform can be seen in a growing concentration on technology
designed to function across the service. New features are developed for
applications like Exchange but an increasing effort is dedicated to functionality
that draws upon multiple parts of Office 365 such as Office 365 Groups, the
Security and Compliance Center, and Delve or support functions like Unified
Auditing. The provision of a suite of REST-based APIs gives programmers a
consistent method to access and exploit various forms of Office 365 data,
including the signals representing user activities that are accumulated in the
Microsoft Graph.

Wave 16: Now in production

The Wave 16 set of the on-premises Office products first shipped to customers in
October 2015. This wave includes Exchange 2016 and SharePoint 2016 as well
as the Office 2016 desktop suite. Although the new generation of server products
underpinned increased reliability and robustness in the base workloads, the
appearance of “cloud-only” applications is the most interesting development
during this period. Teams, Planner, Flow, PowerApps, Sway, and StaffHub are
some of the examples of major change within Office 365 that will never appear
on-premises.
During this time, Microsoft consolidated its cloud properties to bring
Outlook.com to the Exchange Online infrastructure. The consumer and business
email applications share the same servers, storage, and software base with
functionality delivered to end users being controlled by different license tiers.
OneDrive and OneDrive for Business also share many components. A good
example fo the value of shared infrastructures is seen in the transfer of
functionality between consumer and business applications, such as OneDrive’s
Restore Files feature and Exchange’s Encrypt email feature, both of which are
available across consumer and business platforms.
Another major change is the leaving behind of workload-specific compliance.
Where SharePoint had the eDiscovery Center and Exchange had its eDiscovery
searches, the Office 365 Security and Compliance Center became the focus for
platform-wide compliance functionality. Not only does the new technology work
for the base workloads and new applications, it introduces new features like
manual disposition, event-based retention, and supervisory policies.
The Security and Compliance Center has taken over some functionality
previously managed by Exchange like anti-malware. This is also part of a trend
to move functionality away from workload administrative portals to Office 365
administration portals. You still need to use the Exchange Administration Center
or SharePoint Administration Center (now refreshed), but not as often as you
used to, and you will visit them less in the future.
Finally, this wave marks the beginning of the transition from Skype for Business
Online to Teams and the Microsoft Phone System. The roots for Skype for
Business Online are in a long line of on-premises servers. The new voice and
video platform is cloud-only and shared with the Skype consumer application.
The new platform is more adaptable to the changing needs of businesses for
voice and video communications, but the transition to Teams will take some
years yet.

The Next Wave

With so much happening in Wave 16, speculation inevitably turns to what might
happen in the next wave of product development. For on-premises customers,
the answer lies in the 2019 generation of the Exchange, SharePoint, and Skype
for Business servers. These products remain good at what they do and deliver
good value to customers if they simply want an email server, a document
management server, or a communications server.
The difference in the cloud is integration. Office 365 is a fabric that gives
development groups a toolset to build new applications and functionality around.
Teams is a great example of how to bring components from across Microsoft’s
cloud properties together to create a new application. The intermingling of
components from different places within Office 365 is a trend we can expect to
continue.
The role of Exchange and SharePoint, the two basic workloads in Office 365,
has diminished in some respects as Office 365 evolved. Exchange was the king
in the early days of Office 365 because email was the first and easiest workload
to move to the cloud. SharePoint followed as migration tools matured and
customers became more used to the idea of managing documents in the cloud.
Both applications came from a position on-premises where they sat at the center
of ecosystems, surrounded by people and other applications. It is very different
inside Office 365.
Today, Exchange is no more than an email server for Office 365 that happens to
provide a convenient way to store some information (like Teams compliance
records). SharePoint manages documents for other applications and itself. The
focus is no longer on Exchange or SharePoint; it has shifted to the integration of
the base workloads into other applications. Thus, we see Exchange deliver
shared mailboxes to Teams and Groups, and SharePoint give Teams and Planner
a convenient place to store documents uploaded to these applications.
Exchange and SharePoint are still extraordinarily important to Office 365 and
every Office 365 administrator should understand how to manage these
applications. This need will continue, but might become less critical as time goes
by as Microsoft automates and simplifies cloud operations. There’s lots more to
learn and master in your Office 365 journey, including:

Azure Active Directory (basic operations, plus extended functionality like

conditional access policies).
Azure Information Protection.
Teams.
Planner.
Flow and PowerApps.
Enterprise Mobility and Security, including InTune.
PowerShell and the Microsoft Graph to automate/script operations.

When we set out to write the first edition of this book in 2014, none of the topics
listed above apart from PowerShell were covered. Now, they’re fundamental
parts of the Office 365 landscape. It’s enough to keep everyone busy.

Microsoft 365
From a business perspective, the bundling of Office 365 into Microsoft 365 is
the most important influence on Office 365 for the immediate future. Microsoft
has invested heavily in cloud infrastructure to build out its datacenters and
networks to support Office 365 and Azure. The need exists to achieve a return on
that investment, and that means that Microsoft must continue to grow the
number of paid subscriptions for Office 365 and increase the annual revenue for
each subscription. Growing the revenue per seat is done by convincing
customers to upgrade their subscriptions to a higher-priced plan (from E3 to E5,
for example) or by buying add-ons for specific functionality. Many of the new
features being added to Office 365 now require E5 licenses and a growing gap is
developing between the functionality available to E3 and E5 tenants to justify
the extra cost of E5 licenses.
Convincing Office 365 customers to embrace Microsoft 365 is another example
of driving extra revenue, and to support the activity, you’ll see that Microsoft
constantly emphasizes the value to customers of deploying Office 365,
Enterprise Mobility and Security, and Windows 10 together. Marketing and
engineering support to illustrate the benefits of Microsoft 365 will appear in a
continual flow to convince customers to embrace the program. If you want to
continue using Office 365 on its own, you can, but a time might come when all
you can buy is Microsoft 365.
Chapter 2: Exchange Mailbox Migration
Paul Cunningham

Office 365 is an attractive IT platform for new businesses because it avoids the
time and expense of purchasing and deploying an on-premises infrastructure. A
green field deployment of the core services of Office 365 is straightforward
because the business is not burdened by any legacy infrastructure or data. On the
other hand, organizations that have an on-premises infrastructure usually need a
certain amount of planning before they can migrate anything to Office 365. For
most customers, email is the first workload to move to the cloud, because the
migration methodology and tools are well established and flexible enough to
meet the requirements of almost any migration scenario. This chapter explores
the various migration methods available from Microsoft and third parties to
migrate email to Exchange Online.
Identity management is a key component of any Office 365 deployment. Many
questions need to be answered about how identity will be managed during and
after the migration. Of course, identity management is important from a security
perspective, but it is also important to consider how it impacts the end user
experience. Chapter 3 (main book) examines the different identity models
available for Office 365. You should understand the material presented there
before you choose a migration method. Take your time on these matters. It is
possible that a specific identity model will cause you problems with your
preferred migration method. Or you might choose a migration method, reach the
end of your migration project, and discover that your ongoing identity needs
have not been met. Be prepared to be flexible in your decision making, and if in
doubt, always consider the user experience implied in your chosen approach. A
poor user experience means a poor perception of the project outcome, even if the
technical execution of the project goes well.
This chapter examines the decision-making process for choosing a migration
method, reviews the cutover and staged migration processes, and provides an
overview of hybrid configurations and other non-Exchange migration methods.
Hybrid configurations are also covered in much more detail in chapter 4. You
may well find that Hybrid is the best approach for your organization, but it is
still well worth your time to read and understand the other options that are
available, so that your decision is an informed one. Let’s begin with a look at the
different migration approaches for Exchange Online.

Migration approaches for Exchange Online

Office 365 supports a variety of migration methods. The choice of migration
method is often influenced by a wide range of factors such as the chosen identity
model, the number of objects (e.g. mailboxes, contacts, public folders) involved
in the migration, the amount of data to be moved to Office 365, the version of
Exchange (if any) running on-premises, long-term migration or co-existence
requirements, whether the organization uses non-Exchange email servers, and
even the budget available to spend on the migration project.
The migration methods that are available can be summarized as:

Cutover migration.
Staged migration.
Hybrid configuration.
PST-based migration.
IMAP migration.
Third party migration tools.

The best place to start is with the business requirements for the migration
project. Business requirements should include factors such as the need to
complete the migration by a particular date, whether a back-out option for the
migration needs to be included, or if some email workload will remain on-
premises. As you will see, each migration method has different benefits and
constraints, and they may not all suit the business requirements of the project.
Technical requirements are considered next. These often eliminate some of the
migration methods and allow the organization to zero in on the feasible
approaches. Figure 2-1 provides an example of the decision-making process you
can work through based on your technical requirements to understand the
available migration methods for your scenario. Even if you find that you meet
the technical requirements of a migration method, you should continue to
research the actual processes involved in performing that migration, because you
might still discover some undesirable element that steers you in another
direction.
Figure 2-1: Decision tree for choosing a migration method
The decision-making process begins by determining which version(s) of
Exchange exist in the environment (if any) because the version(s) of Exchange
in use can narrow the available migration methods, or at least those provided by
Microsoft. A specific question is whether Exchange 2003 or 2007 servers exist
within the organization. These are now very old servers, so it is unsurprising that
the need to migrate data from these servers would limit the available options.
Hosted Exchange providers complicate matters further, because the hosting
provider usually limits a customer from performing the types of configuration
and preparation that are required for built-in migration methods. Unless the
Hosted Exchange provider is very cooperative, a third-party migration tool
might be needed to migrate from a Hosted Exchange service to Office 365. That
said, third party migration tools have their own set of requirements and
limitations, which will vary depending on the product, not to mention the
additional cost involved, which needs to be factored in to your project budget.
Table 2-1 summarizes the built-in migration methods available for different
versions of on-premises Exchange. The Exchange versions listed refer to the
highest version of Exchange in the organization. For example, if an organization
runs a mixed Exchange Server 2010 and 2007 organization, then they can use
the migration methods supported for Exchange 2010 and are not limited to the
options available for Exchange 2007. Some additional requirements and
constraints are mentioned in Table 2-1 that will be explained in more detail later.
Exchange
Cutover Staged Hybrid IMAP
Version
Exchange Yes (if under No (unless a Hybrid server running
Yes Yes
2003 2,000 mailboxes) Exchange 2010 is deployed)
Exchange Yes (if under No (unless a Hybrid server running
Yes Yes
2007 2,000 mailboxes) Exchange 2010 or 2013 is deployed)
Exchange Yes (if under
No Yes Yes
2010 2,000 mailboxes)
Exchange Yes (if under
No Yes Yes
2013 2,000 mailboxes)
Exchange Yes (if under
No Yes Yes
2016 2,000 mailboxes)
Table 2-1: Available migration methods for on-premises Exchange versions
In addition to the built-in migration methods, you can consider:

Migrating user PSTs using the Office 365 Import Service, Microsoft’s PST
Collection tool, or a third-party migration tool. PST-based migrations are
discussed later in this chapter.
Third party migration tools that use protocols like Exchange Web Services
(EWS) to ingest data into Exchange Online mailboxes. Third party tools
often provide solutions to very complex migration scenarios that built-in
 migration methods cannot handle. A list of third party migration vendors is
included later in this chapter.

Note: Although IMAP is included in Table 2-1, it is the least preferable

migration method when you migrate from an Exchange server, and is generally
only suitable when migrating from non-Exchange platforms such as Gmail or
Yahoo!. We discuss the limitations of IMAP migrations later in this chapter.
The next consideration is whether there are more than 2,000 mailboxes.
Organizations with fewer than 2,000 mailboxes are supported for cutover, staged
and hybrid migrations, while organizations with more than 2,000 mailboxes are
only supported for staged or hybrid migrations. The 2,000-mailbox limit does
not mean that organizations with less than 2,000 mailboxes should automatically
choose a cutover migration. For example, if the organization wants to migrate
their users in smaller batches instead of one big batch then a cutover migration is
not suitable.
Real World: 2,000 mailboxes is the threshold specified by Microsoft in terms
of support for cutover migrations. The logistics involved in handling an outage
for such a large number of users, as well as the desk-side support needed to
assist with reconfiguring Outlook profiles and mobile devices after the cutover,
may simply make a cutover migration too risky and complex for the
organization. In fact, many experienced Office 365 consultants consider the
practical limits of both the cutover and staged migration methods to be as few
as 150 mailboxes. Organizations larger than 150 mailboxes should give strong
consideration to using a hybrid migration instead of a cutover or staged
migration.
When cutover is either not possible or not desirable for an Exchange 2003/2007
organization, the remaining options are staged and hybrid migrations. For an
Exchange 2003 environment an Exchange 2010 server can be deployed to create
a Hybrid configuration. If you want to use an Exchange 2013 or Exchange 2016
server to host the hybrid configuration, you will have to complete a full
migration to Exchange 2010 first. For an Exchange 2007 organization at least
one server running either Exchange 2010 or Exchange 2013 must be installed to
provide the hybrid functionality. Both staged and hybrid options require the
implementation of directory synchronization. Without directory synchronization,
your migration options are limited to the use of third party migration tools.
An organization migrating from Exchange 2003 or Exchange 2007 can choose to
take advantage of the free Hybrid license available from Microsoft. This license
allows an Exchange 2016, 2013 or Exchange 2010 SP3 server to be deployed in
the organization to facilitate a hybrid connection with Office 365 (depending on
the supported versions that can co-exist in an organization). The Hybrid license
can’t be used for a server that hosts mailboxes, but the server can be used during
the migration to Office 365 and retained afterwards to manage the Exchange
attributes of the on-premises Active Directory objects. A server assigned with a
Hybrid license can also be used as an SMTP relay server for applications or
devices on the corporate network.
If the implementation of a Hybrid server is not possible, for example due to
server capacity constraints, then a staged migration is the way forward. The
staged migration method is not available for organizations that run Exchange
2010 or later. Exchange 2010 or later environments with fewer than 2,000
mailboxes to migrate can still choose to perform a cutover migration. However,
as we’ve already discussed, large cutover migration projects can be logistically
very difficult to perform.
Given that Exchange 2010 (or later versions) is capable of hybrid configuration
with Office 365 you should give strong consideration to using the hybrid
approach instead of a staged or cutover migration. Although hybrid is the most
complex of all the migration options in terms of initial setup and configuration, it
delivers the best user experience during the migration. Hybrid configurations
allow the on-premises Exchange organization and Office 365 to function as
though they are the same environment with seamless mail flow, a shared address
book and calendar free/busy federation. In fact, most users would not even be
aware that they are working in a hybrid configuration with mailboxes deployed
in both on-premises and Office 365. A hybrid configuration is also the only
option that allows mailboxes to be off-boarded from Office 365 to Exchange on-
premises without using third party migration tools. Cutover and staged
migrations can’t off-board or roll back to an on-premises server without
significant effort and the risk of data loss.
Hybrid configurations require directory synchronization of the on-premises
Active Directory objects into Azure Active Directory, so that they can be used in
Office 365. If for some reason the organization can’t implement directory
synchronization, then the choices are limited to third party migration tools.
Finally, businesses using non-Exchange email platforms such as Gmail or
Google Apps for Business can’t use the cutover, staged or hybrid options. For
those businesses Microsoft provides the IMAP migration option to move
mailboxes to Office 365. Alternatively, a third-party migration tool can be used.
If you’re engaging the Microsoft Onboarding Center or an outside consultant to
assist with the migration, they will usually recommend a specific tool or method
for the migration, which will probably be the tool for which they have most
experience.
All aspects of the decision-making process require careful consideration. Beyond
the technical considerations are also other factors such as whether the migration
project will be handled in-house or by an external consultant, whether extra
training is required for IT staff to understand new features such as hybrid
configurations, and whether funding is available to pay for third party migration
tools if built-in migration options can’t be used.
If you want some specific recommendations to get you started, it is generally
recommended to use:

Hybrid configuration for Exchange 2010 or later.

PST-based migration if a non-Exchange email system can extract data to
PST files.
IMAP or third-party tools for any non-Exchange email systems that can’t
extract data to PST files.
Third party solutions for very complex migration scenarios.

Note: Before you finalize your decision on which migration method to use it is
strongly recommended that you read through the example migration scenarios
from start to finish, so that you can learn about any risks or timing issues that
you need to be prepared for. Do not start a migration before you have read
through the process from start to finish at least once, and you understand the
support implications of your migration method and ongoing identity
management. You should also consider creating a test environment and signing
up for a separate Office 365 trial tenant so that you can perform a hands-on test
run of your chosen migration method.

Managing a Migration Project

Every migration project needs some degree of project management ranging from
a large project with a dedicated project management team to a small project that
you self-manage. Before you launch into the details of figuring out the various
configuration and migration tasks it is wise to start with a project planning
exercise.
During the planning phase, you should hold a kick-off meeting with the
stakeholders of the project. This will ensure that everyone has a chance to
discuss what the Office 365 migration will mean for them, identify any key
success criteria, and to flag any concerns or potential issues that they foresee
with the migration. It’s always helpful to know what the stakeholders’ biggest
worries are so that you can address them with a technical solution or simply by
providing more communication about that matter.
The planning phase is also the opportunity to collect information about the
current environment such as the size and number of user mailboxes, shared
mailboxes, public folders, other mail-enabled objects such as contacts and
distribution groups, delegates, and any applications or systems that rely on
email, and network connectivity for the sites involved. If your migration is for
more workloads than just email, then you can also collect information about file
shares, SharePoint sites, or Skype for Business configurations that will be
needed. The information collected during this stage can also feed into the
decision-making process for choosing an identity model and migration method.
After your planning is complete, you should begin communicating with end
users about what they can expect from the migration, and give them some
advanced warning about anything they will need to do as part of the migration.
Good communication can be the difference between a successful project and one
that is considered unsuccessful. A flawless technical migration will still attract
criticism from end users if they are surprised by outages during the migration or
different user experiences afterwards.
Develop a checklist of migration tasks to follow throughout your migration
project. This will help you to avoid missing any crucial steps that might slip your
mind during the busy parts of the migration. Every organization’s checklist will
be different in some way, and if you’re doing multiple migration projects for
different customers you will develop a good checklist of your own over time.
Microsoft also publishes a checklist for Office 365 deployments that can be used
as a starting point for creating your own.
Finally, ensure that you create a test plan based on the information that you
collected during the planning phase. A good test plan describes how services are
used in your organization, beyond their basic functionality. For example, testing
that email works when sending to or from external addresses confirms that the
basic functionality works, while testing that email notifications from your CRM
work correctly is a test that is relevant to your specific organization. The more
scenarios you can think of for your organization that are more than just basic
functionality, the more robust your test plan will be. You can use this test plan
during the migration project to ensure that important business systems continue
to work correctly, and to verify that end users will have the experience they are
expecting post-migration.

Preparing to Migrate to Exchange Online

Before you can migrate to Office 365, you need to sign up for a tenant. At this
stage, you can opt to sign up for a free trial that will run for 30 days, or you can
decide to start paying immediately. The free trial period has full functionality,
and is a chance to try out features that you are unsure whether you will need. For
example, you can sign up for an Enterprise E5 trial, but then you might discover
during the first 30 days that you only need E1 or E3 licenses instead.
If you are unsure of your ability to work with Office 365 and need to gain
experience, it’s best to create a separate trial tenant at the start and use it to
experiment with different services and settings. Then, after you’ve acquired
sufficient experience, you can create a tenant that you intend to use for
production services, and begin to prepare it for the migration of on-premises
workload.
During the signup process, you’ll be asked where your organization is located.
Selecting the right country for your organization will determine where your
Office 365 tenant is located around the globe. For organizations that span
multiple geographic regions you should choose the country where most of your
end users will be located. In Chapter 1 you learned that Microsoft operates many
datacenters in different regions to provide Office 365 service to end users. Your
proximity to your datacenters is a factor in the quality of the user experience for
your Office 365 services, but not a defining one. A more important consideration
for organizations that reside in countries with strict data sovereignty regulations,
for example Germany, is ensuring that the Microsoft datacenters hosting your
services are the appropriate ones.
As you move through the signup process you will be asked to create the first
administrator account for your tenant. This step also involves choosing the
service domain (or tenant name) for your tenant, which is the
*.onmicrosoft.com domain assigned to every Office 365 tenant. The service
domain will be visible to your end users in the SharePoint Online URL, Skype
for Business meeting invites, and OneDrive for Business libraries, so choose a
name that aligns with your organization’s company name or brand. The service
domain name must be unique within Office 365. Unfortunately, if another tenant
already has the name that you would like to use there is no way to get the name
for yourself.
The service domain also cannot be changed later. Some customers make the
unfortunate error of signing up for a tenant with a service domain such as
“contosotest.onmicrosoft.com”, then begin using it for production services
before they realize it cannot be changed. A similar naming problem can occur
when the company goes through a rebranding, or is acquired by another
company. If it is important to your organization to change tenant names, the only
options are to perform an off-boarding migration to on-premises servers, then
migrate to a new Office 365 tenant, or alternatively to perform a tenant-to-tenant
migration. Both options involve considerable effort and cost, and are best
avoided unless necessary. It's possible that in future Microsoft will develop a
tenant renaming process that makes the situation easier for customers, but
considering the complexity of Office 365 and the number of integrated services
involved it could be quite some time before that capability arrives.
Real World: Companies that operate with multiple parent-child companies and
different brand identities have some challenges with Office 365 tenant naming
to consider. If the companies do not share an email domain or any other data
and resources, then it’s feasible to consider separate Office 365 tenants.
However, if there are shared domains and resources, and a single Office 365
tenant is deemed necessary, then a common approach is to use a single parent
company as the tenant name, such as “contosoholdings.onmicrosoft.com”.

Adding Domain Names

Migrating to Exchange Online usually means moving across an existing email
domain to Office 365, but some migration scenarios can also involve migrating
to a new domain name, for example when a portion of a company is divested and
moves to its own corporate brand. In either case, the domain names that will be
used by email recipients need to be added to Office 365. Exchange
administrators will be familiar with this requirement from managing Accepted
Domains in on-premises Exchange environments. For Office 365 the domain
names are used by multiple services, not just by Exchange Online, so they are
managed through the Office 365 admin center and are referred to as “vanity
domains”, or simply as “domains”. A domain name can only be verified in one
Office 365 tenant at a time.
Domains are added to a tenant by logging in to the Office 365 admin center and
navigating to Settings and then Domains. Click the Add domain button to start
the wizard that will guide you through the process. As part of this process you’ll
need to validate the domain by adding a record to the DNS zone to prove your
ownership and control. There are two validation records offered; an MX record,
and a TXT record. Adding an MX record that points to Office 365 at this early
stage of a migration is likely to cause disruption to your existing mail flow, so I
recommend that you use the TXT record method instead.
Microsoft can host the DNS for your Office 365 domains. You have the option to
use Office 365 DNS services, or continuing to manage your own DNS records
with your existing host. For domains that are already in use by an organization
there are no immediate advantages to moving the DNS hosting to Office 365, but
it’s an option you might consider if you are adding a brand-new domain to
Office 365 in the future, or if you want to move all aspects of your messaging
service to Office 365.
As additional tasks after adding domain names to your Office 365 tenant, you
can consider:

Adding administrator user accounts. Having multiple administrators in

Office 365 is quite common, especially in larger IT organizations. As
explained in Chapter 5 (main book), you can use different administrator
roles to control the access that administrators have.
Adding user accounts. This will depend on the identity model and migration
approach you are using. For example, the cutover, staged, and hybrid
migration approaches all have specific steps for provisioning users in Office
365, while other methods such as third-party tools might require you to
manually provision the user accounts. You should check the requirements
of your chosen migration method first, before you create any accounts, so
that you don’t cause errors and unnecessary rework later.

Configuring the On-Premises Infrastructure

The migration service for cutover and staged migration methods uses Outlook
Anywhere to connect to an on-premises Exchange server and synchronize the
mailbox contents. Outlook Anywhere (also called RPC-over-HTTP in Exchange
2003) is not enabled by default on Exchange 2010 or earlier, so you need to
review your server configuration and enable Outlook Anywhere. A valid third-
party SSL certificate will also be required for your server, if one is not already
installed. Self-signed certificates simply won’t work. This is a nominal cost from
most certificate authorities such as Digicert. You do not need to spend thousands
of dollars on a new, high-end SSL certificate just to satisfy the requirements of a
migration to Exchange Online.
Outlook Anywhere requires TCP port 443 (HTTPS) to be open on your firewall
and NATing or port forwarding to the on-premises Exchange server. The hybrid
migration method uses Exchange Web Services (EWS), which also operates over
HTTPS. EWS is used by several third-party tools as well, but you should always
check the documentation provided by those vendors for specific firewall
requirements. If you have not already opened firewall access for services such as
Outlook Web Access and ActiveSync, then you should review your firewall
configuration at this stage and make any necessary changes to allow the HTTPS
connections.
The Exchange Remote Connectivity Analyzer (ExRCA) can be used to test the
connection to Outlook Anywhere or Exchange Web Services so that you can
verify that your on-premises infrastructure is configured correctly before you
attempt any further migration steps.
Some organizations that do not already have HTTPS open for external
connections to an on-premises Exchange server may resist the notion of opening
firewall ports to the entire world. In such cases the firewall access can be
restricted to the IP address ranges for Office 365. Using IP address filtering may
be an acceptable temporary measure while the migration is performed, but it
becomes difficult to maintain this configuration over a long period due to the
rate of change that occurs with the Office 365 IP address ranges. Firewalls that
can only filter based on IP addresses may require changes as often as weekly.
Firewalls that can filter based on FQDNs, or DNS names, will require far less
maintenance on an ongoing basis.
Creating Migration Service Accounts
For cutover and staged migrations, the Office 365 migration service needs a set
of user credentials to connect to your on-premises organization and access
mailboxes. It is recommended to create a dedicated service account for this
purpose. Service account requirements for Hybrid scenarios are covered in
chapter 4.
The first step is to create a new service account in the on-premises Active
Directory with a meaningful name, for example “O365Migration”. Make sure
that this account has a strong password, and set the password to never expire so
that you can avoid problems if you have a password policy in Active Directory
that would expire the password before the migration is complete.
One method suggested by Microsoft’s TechNet documentation to grant the
service account the necessary permissions is to add it to the Domain Admins
group in Active Directory. However, this makes the account very powerful and
increases the risks if the credentials are compromised, so it is not recommended.
A lower risk approach that aligns with best practices is to grant the user
permissions only to the mailboxes in Exchange. You can perform this on each
mailbox, or on each mailbox database. The advantage of doing it at the database
level is that any new mailboxes created after the permissions are granted will
automatically inherit the required permissions, while the per-mailbox method
requires you to manually add the permissions to any new mailboxes created
later.
To grant the service account permissions for a mailbox database, you can run
this command.
[PS] C:\> Get-MailboxDatabase | Add-ADPermission -User NRU\O365Migration
-ExtendedRights Receive-As

Applying the permissions at the database level is simpler, however, if you decide
to grant the service account permissions for each mailbox instead you can run a
single PowerShell command.
[PS] C:\> Get-Mailbox | Add-MailboxPermission -User NRU\O365Migration
-AccessRights FullAccess

Note: If you are migrating from Exchange 2003, PowerShell is not available to
run the commands shown above. Instead, you can use the Exchange System
Manager console to add the permissions to each database.

Reducing the Migration Load

The more data that is migrated to Exchange Online the longer it will take. To
enable a speedier migration, you can undertake a clean-up process to reduce the
overall size of mailbox data that will need to be migrated. Emptying deleted
items from mailboxes is a quick win, and in some customer environments I’ve
worked in this has reduced the overall load by as much as 20%. Aged data is
also a prime candidate for clean-up. If emails older than a certain number of
years are no longer required, then retention policies can be used to remove them
from the on-premises mailboxes ahead of the planned migration.
One of the most common causes of skipped items during a migration is the size
of the item itself. Office 365 has an advertised, default per-item size limit of 25
MB, which amounts to approximately 35 MB once the combined size of the
message contents, file attachments, and other metadata are considered.
Administrators can increase this up to 150 MB (see Chapter 17 of the main
book) to allow larger email attachments to be sent and received.
Another random document with
no related content on Scribd:
comprehensiveness, and again to the inner harmony or systematic
structure of the interest it embodies, it constitutes a genuine self-
existing individual whole of the kind which psychologists recognise
as a “self.” And again, in so far as my life exhibits determinate
character, so far do these systematic purposes or minor “selves”
form a larger system, also individual, which may be called my “total
self.” And both the many lesser “selves” and the larger “self” are real
in the same sense of the word. Neither exists merely in or for the
other; the wider or whole “self” is no mere collection or resultant or
product of the more special “selves,” nor are they again mere results
of a theoretical process of analysis and abstraction. In so far as they
are genuine systems at all, they are not mere “parts” of a whole, but
each is the expression, in a concrete conscious life, of the nature of
a larger whole from a special “point of view.” The whole, if not
equally in every part, is yet as a whole present in every part, and
precisely for that reason the category of part and whole is
inadequate to express their relation. Somewhat after this fashion we
must conceive the structure of any individual whole of lesser
individuals. Why, in spite of the analogy, it is desirable not to speak
of the whole of Reality as a “Self,” will be made clearer as we
proceed.[63]
§ 8. The view we have formulated is perhaps more closely akin to
Spinoza’s conception of the relation of the human mind to the
“infinite intellect of God,” than to any other historically famous theory.
According to Spinoza, the individual human mind is an “eternal mode
of consciousness which, taken together with all other such ‘modes,’
makes up the infinite intellect of God.” The meaning of the epithet
“eternal” we cannot, of course, enter into until we have discussed the
relation of the time-process to experience. The rest of the definition
pretty clearly coincides in its general sense with the view we have
tried to expound of the nature of the relation between the supreme
experience and its constituent experiences. For the “modes” of
Spinoza are definitely thought of as genuinely individual
manifestations of the nature of his ultimate reality, “substance” or
“God.” Their individuality and their infinite multiplicity is no result of
illusion or illegitimate abstraction. And, on the other hand,
“substance” itself is genuinely individual; it is no mere abstract name
for the common properties of a number of ultimately independent
things.
Most of the adverse criticism which Spinozism has met with, as far
at least as regards its doctrine of the nature of the human mind,
seems to be based on misapprehension about the first of these
points. From his use of the numerical category of whole and part to
express the relation between substance and its modes, Spinoza has
incorrectly been taken to be denying the fact of the genuine
individuality of the finite experience, and therefore to be declaring the
very existence of the finite to be mere baseless illusion. With his
doctrine as thus misinterpreted, ours has, of course, no similarity.
Nothing is explained away by calling it “illusion”; the “illusory” fact is
there in spite of the hard names you choose to bestow on it, and
demands explanation no less than any other fact. Our theory aims
not at dismissing finite individuality as illusion, but at ascertaining
what it means, what are its limits, and how it stands related to the
complete individual whole of experience which Spinoza calls the
infinitus intellectus Dei.[64]
The mention of Spinoza will no doubt suggest to the reader the
famous doctrine, which has played so large a part in the subsequent
development of philosophical Monism, of the double “aspects” or
“attributes” of Reality. It is from Spinoza that modern Monism has
learned the view that the mental and physical orders are related as
two parallel but distinct manifestations of a common underlying
reality, so that to every member of one order there corresponds a
determinate member of the other. The two are thus everywhere
inseparable and everywhere irreducible “parallel” expressions of a
nature which is neither mental nor physical. On this fundamental
point our theory, as will have been seen already, completely parts
company with Spinozism. That the nature of one and the same
common whole should be equally manifested in two entirely
irreducible forms, is a patent impossibility. Either the unity of the
whole or the absolute disparateness of its twin manifestations must
be surrendered if we are to think consistently. Hence we cannot
avoid asking in which of the two series the assumed common nature
is more adequately expressed. According as we answer this
question we shall find ourselves led in the end either to thorough-
going Materialism or to thorough-going Idealism. For our own part,
the perception that Reality is experience and nothing else has
already committed us to the view that both of the seemingly
disparate series must in the end be mental. Thus our doctrine may
be said to be much what Spinoza’s would be if the attribute of
“extension” were removed from his scheme, and the whole of Reality
identified with the “infinite intellect of God.”[65]

Consult further:—B. Bosanquet, Essentials of Logic, lect. 2; Logic,

vol. ii. chap. 7; F. H. Bradley, Appearance and Reality, chaps. 13, 14,
20; L. T. Hobhouse, Theory of Knowledge, pt. 3, chap. 6, “Reality as
a System”; H. Lotze, Metaphysic, bk. i. chap. 6 (Eng. trans., vol. i.
pp. 163-191); J. S. Mackenzie, Outlines of Metaphysic, bk. i. chaps.
2, 3; bk. iii. chap. 6; J. E. M‘Taggart, Studies in Hegelian Cosmology,
chap. 2.
 54. This consideration obviously influenced Leibnitz. It is a much-
decried doctrine in his system that every “monad,” or simple real
thing, perceives nothing but its own internal states; there are no
“windows” through which one monad can behold the states of
another. It is easy to show that this doctrine leads to extremely far-
fetched and fantastic hypotheses to account for the apparent
communication between different monads, but not so easy to show
that Pluralism can afford to dispense with it. See in particular
Leibnitz’s New System of the Nature of Substances (Works, ed.
Erdmann, 124 ff.; ed. Gerhardt, iv. 477 ff.; Eng. trans. in Latta’s
Leibnis: the Monadology, etc., p. 297 ff.), especially §§ 13-17 and
Monadology, §§ 7-9, 51.
55. See for a recent treatment of this point in its bearing upon the
theory of volition and moral accountability, Mr. Bradley’s article on
“Mental Conflict and Imputation” in Mind for July 1902. There is
probably no part of Psychology which suffers more from an improper
over-simplification of awkward facts.
56. As the reader will readily collect from the preceding discussion,
I do not myself admit that they are justified. On the contrary, I should
hold that any consistent Pluralism must issue in what, if I held it
myself, I should feel compelled to describe as Atheism, and the
doctrine of blind chance as the arbiter of all things. In this matter I
should like to associate myself entirely with the emphatic protest of
Mr. Bradley, in Mind July 1902, p. 313, and with the remarks of Mr. B.
Russell in his work on The Philosophy of Leibniz, p. 172. I need not
say that I do not make these remarks for the purpose of
disparagement. By all means, if Atheism is the logical outcome of
consistent thinking, let us say so; what I object to is the constant
appeal to theistic beliefs on the part of metaphysicians who, so far as
I can see, ought to be atheists if they were in earnest with their own
position.
57. For a popular exemplification of the kind of appeal to religious
and ethical interests here objected to, see the first essay in Prof.
James’s Will to Believe. I have never been able to understand why
these appeals, if legitimate, should not be allowed in Psychology or
any other science as readily as in Metaphysics. Would Prof. James
regard it as a valid argument for the “timeless self” or the
Innervationsgefühle, that some men may be better or happier for
believing in them? Or again, is it in itself an objection to the study of
Ethics that certain persons may become both less moral and less
happy as a consequence of studying it?
58. N.B.—These possibilities, it must be remembered, though
numerically infinite, are assumed to be qualitatively determinate,
being constituted of the condition of conformity to the logical principle
of non-contradiction. Now there is no reason in the nature of a
plurality of independent things why this principle should be
recognised rather than not.
59. A medley of independent things would not even be really
“many.” For until you can count “first, second, third....” you have not
your Many. And nothing but the terms of a coherent connected series
can be counted. What you can count as many is shown by the very
fact of your ability to count it to have a common nature or ground
which permits of its orderly arrangement, and thus to be part of one
system. Compare Plato, Parmenides, 164, 165.
60. As Aristotle more than once says, a human hand, for instance,
is not when severed from the rest of the body a “hand” at all, except
ὁμωνύμως “equivocally,” any more than the “hand” of a statue is a
true hand. (I.e. it is only a “true hand” so long as it does the work of a
hand. Captain Cuttle’s hook probably deserved the name of “hand”
better than the severed member it replaced).
61. I shall attempt to show in a later chapter (Bk. IV. chap. 3) that,
in any useful signification of the term “self,” Reality is not a “self” nor
yet a mere community of “selves.”
62. Again, I must remind the reader that this recognition of the
teleological character of mind does not in the least preclude the
necessity for psychological analysis of mental states. Still less does
it require us to include in our analysis a volitional element as one
distinguishable aspect or component of the isolated mental state by
the side of others, such as the presentational and emotional aspects.
It might even be contended that a “tripartite” or three-aspect
Psychology commits the mistake of counting in the whole psychical
fact as one of its own components.
63. See infra, Bk. IV. chap. 3, where we shall find that the relation
of the individual self to a social whole probably furnishes a still better,
though not altogether satisfactory, illustration of our principle.
64. For Spinoza’s doctrine see especially Ethics, I. 15, 25; II. 11,
40; III. 6-9; V. 22, 23, with the explanations of any good exposition of
his system, such as that of Pollock or Joachim.
65. See further on the “Parallelistic” doctrine, Bk. IV. chap. 2.
 CHAPTER III

REALITY AND ITS APPEARANCES—THE

DEGREES OF REALITY
§ 1. Reality being a single systematic whole, the nature of its constituent elements
is only finally intelligible in the light of the whole system. Hence each of its
“appearances,” if considered as a whole in itself, must be more or less
contradictory. § 2. But some “appearances” exhibit the structure of the whole
more adequately than others, and have therefore a higher degree of reality. §
3. This conception of degree of reality may be illustrated by comparison with
the successive orders of infinites and infinitesimals in Mathematics. It would
be the task of a complete Philosophy to assign the contents of the world to
their proper place in the series of “orders” of reality. § 4. In general any
subordinate whole is real in proportion as it is a self-contained whole. And it is
a self-contained whole in proportion as it is (a) comprehensive, (b) systematic;
that is, a thing is real just so far as it is truly individual. § 5. The two criteria of
individuality, though ultimately coincident, tend in particular cases to fall apart
for our insight, owing to the limitation of human knowledge. § 6. Ultimately
only the whole system of experience is completely individual, all other
individuality is approximate. § 7. In other words, the whole system of
experience is an infinite individual, all subordinate individuality is finite.
Comparison of this position with the doctrines of Leibnitz. § 8. Recapitulatory
statement of the relation of Reality to its Appearances.

§ 1. Reality, we have seen, is to be thought of as a systematic

whole forming a single individual experience, which is composed of
elements or constituents which are in their turn individual
experiences. In each of these constituents the nature of the whole
system manifests itself in a special way. Each of them contributes its
own peculiar content to the whole system, and as the suppression or
change of any one of them would alter the character of the whole, so
it is the nature of the whole which determines the character of each
of its constituents. In this way the whole and its constituent members
are in complete interpenetration and form a perfect systematic unity.
In the happy phrase of Leibnitz, we may say that each of the partial
experiences reflects the whole system from its own peculiar “point of
view.” If we call the completed system, as it is for itself, Reality par
excellence, we may appropriately speak of the partial experiences in
which its nature is diversely manifested as its Appearances. We
must remember, however, that to call them appearances is not to
stamp them as illusory or unreal. They will only be illusory or unreal
when we forget that they are one and all partial aspects or
manifestations of a whole of which none of them adequately
exhausts the contents.
When we forget this and treat any partial experience as though it
were the complete and adequate expression of the whole nature of
Reality,—in other words, when we try to apply to existence or the
universe as a whole conceptions which are only valid for special
aspects of existence,—we shall inevitably find ourselves led to
contradictory and absurd results. Each partial aspect of a total
system can only be ultimately understood by reference to the whole
to which it belongs, and hence any attempt to treat the part in
abstraction as itself a self-contained whole,—or, in other words, to
treat the concepts with which we have to work in dealing with some
special aspect of the world of experience as ultimately valid in their
application to the whole system,—is bound to issue in contradiction.
Again, just because our knowledge of the structure of the system as
a whole is so imperfect as it is, our insight into the structure of its
constituents is also necessarily limited. Hence it will commonly
happen that, even within the limits of their applicability, the special
concepts of our various sciences are not, when thought out, free
from internal contradiction. For instance, we are led to absurd results
when we try, as Materialism does, to interpret the whole system of
experience in terms of the concepts used in the purely physical
sciences; and again, even in their restricted use as physical
categories, these concepts seem incapable of being so defined as to
involve no element of contradiction.
In both these senses all Appearance implies an element of
contradiction; only for an insight which could take in at once the
whole system of existence would its details be completely coherent
and harmonious. But this does not alter the fact that, so far as our
insight into any part of the whole and its connection with other parts
is self-consistent, it does convey genuine, though imperfect,
knowledge of the whole. Though our detailed insight into the
structure of the whole may never reach the ideal of perfect self-
consistency, yet it may approximate to that ideal in different degrees,
at different stages, or with reference to different aspects. And the
closer the approximation the less the modification which our
knowledge would require to bring it into complete harmony with itself,
and the greater therefore the element of truth about Reality which it
contains.
In particular, we must carefully avoid falling into the mistake of
thinking of the Reality and the world of its appearances as though
they formed two distinct realms. In a systematic unity, we must
remember, the whole can exist only in so far as it expresses its
nature in the system of its parts, and again the parts can have no
being except as the whole expresses itself through them. To the
degree to which this condition is departed from by any of the types of
system familiar to us, those systems fall short of being perfectly
systematic. Reality, then, being a systematic whole, can have no
being apart from its appearance, though neither any of them taken
singly, nor yet the sum of them thought of collectively,[66] can exhaust
its contents. And though no appearance is the whole of Reality, in
none of them all does the whole Reality fail to manifest itself as a
whole. The whole is truly, as a whole, present in each and every
part, while yet no part is the whole.[67]
We may once more illustrate by an appeal to our own direct
experience. Consider the way in which we set to work to execute any
systematic scheme or purpose, e.g. the mastery of a particular
science or a particular business. We have in such a case a central
aim or purpose, which in the process of execution spreads out into a
connected system of subordinate ideas and interests welded into
one by the reference to a common end which pervades the whole.
The supreme or central aim is only realised in the successive
realisation of the subordinate stages; at the same time, while it is
what sustains all the members of the system, it has no existence
apart from them, though it is identical neither with any one of them
nor yet with their sum collectively considered.
§ 2. If our conviction that Reality is a single systematic unity
pervading and manifesting itself in lesser systematic unities is
correct, we shall expect to find that some of the lesser systematic
unities with which we have to deal in practical life and in the various
sciences exhibit more of the full character of the whole to which they
belong than others. The “points of view” from which each minor
system reflects the whole, though all true, need not be all equally
true. Though the whole, in a genuine system, must be present as a
whole in every part, it need not be equally present in all; it may well
not be “as full, as perfect in a hair as heart.” To take a concrete
example, a cluster of mass-particles, a machine, a living organism,
and a human mind engaged in the conscious systematic pursuit of
truth, are all to some degree or other systematic unities, and all to
some degree, therefore, repeat the structure of the universal whole
to which they all belong. But it does not follow that all manifest the
structure of that whole with equal adequacy and fulness. Indeed, any
philosophy which admits development as a genuine feature of the
world-process must maintain that they do not, that the nature of the
whole system of Reality is exhibited with infinitely greater adequacy
and clearness in the working of the conscious mind than in the
changes of configuration of the system of mass-particles or even the
vital processes of the physical organism.
In practical life, too, one of our most ineradicable convictions is
that there are degrees of worth which coincide with degrees of the
adequacy with which partial systems exhibit the nature of the larger
wholes to which they belong. For instance, among the different
mental systems which may be called my partial “selves,” there are
some which I call “truer” than others, on the ground that they more
fully reveal my whole character as an individual human being. My
whole character undoubtedly appears in and determines all the
subordinate systems which make up my mental life. Each of them is
the whole character in a special aspect, or as reacting upon a
special system of suggestions, but some of them contain the whole
in a more developed and explicit form than others. I am in one sense
myself wherever I may be and whatever I may be doing, and yet I
am “more myself” in health than in sickness, in the free pursuit of
self-chosen studies than in the forced discharge of uncongenial
tasks imposed on me by the necessity of earning an income.
 We ought, then, to be prepared to find the same state of things
universally in the relation of Reality to its Appearances. In a world
where “higher” and “lower,” “more” and “less” true have a meaning,
some of the lesser systems in which the nature of the whole is
expressed must be fuller and more adequate representations of that
nature than others. This is as much as to say that it would require
comparatively little transformation of some of the partial systems
recognised by our knowledge to show how the common nature of the
whole system of Reality is expressed in them; in other cases the
amount of transformation required to show how the whole repeats
itself in the part would be much more extensive. To take a single
instance, if our preceding analysis of the general nature of Reality is
sound, we can see much more clearly how that nature reappears in
the structure of a human mind than how it is exhibited in what we call
a physical thing, and we may therefore say the human mind
expresses the fundamental character of the whole system much
more fully and adequately than physical nature, as it exists for our
apprehension. More briefly, the same thought may be expressed by
saying that Reality has degrees, and that the forms of Appearance in
which its common nature is most fully and clearly manifested have
the highest degrees of reality.
§ 3. This conception of Reality as capable of degrees may at first
seem paradoxical. How can anything, it will be asked, be more or
less “real” than anything else? Must not anything either be entirely
real or not real at all? But the same difficulty might be raised about
the recognition of degree in other cases where its validity is now
universally admitted. Thus to some minds it has appeared that there
can be no degrees of the infinite or the infinitesimal; all infinites, and
again all zeros, have been declared to be manifestly equal. Yet it
hardly seems possible to escape the conclusion that the concept of
successive orders of infinitely great, and again of infinitely small,
magnitudes is not only intelligible but absolutely necessary if our
thought on quantitative subjects is to be consistent (When the sides
of a rectangle, for instance, become infinitely great or infinitely small
relatively to whatever is our standard of comparison, it still remains a
rectangle, and its area therefore is still determined by the product of
its sides, and is therefore infinitely great or small, as the case may
be, in relation not only to our original standard but to the sides
themselves.[68]) What is in one sense not a matter of degree, may yet
in another not only admit but positively require the distinction of
degrees of more and less. And this is precisely the case with Reality
as it manifests itself in its various appearances. In the sense that it is
the same single experience-system which appears as a whole and in
its whole nature in every one of the subordinate experience-systems,
they are all alike real, and each is as indispensable as every other to
the existence of the whole. In the sense that the whole is more
explicitly present in one than in another, there is an infinity of
possible degrees of reality and unreality. We should be justified in
borrowing a term from mathematical science to mark this double
relation of the appearances to their Reality, and speaking of them as
successive orders of Reality. And we might then say that it is one of
the principal problems of a complete Philosophy to ascertain and
arrange in their proper sequence, as far as the limitations of our
knowledge permit, the orders of Reality.
§ 4. Such a task as this could only be carried out by an intelligence
equally at home in metaphysical analysis and in the results of the
special sciences, and would form the proper work of applied
Metaphysics. In a discussion of general metaphysical principles it is
sufficient to indicate the general nature of the criteria by which the
degree of reality exhibited by any special partial system must be
determined. Now, this general nature has been already made fairly
clear by the foregoing inquiry into the unity of Reality. Reality, we
have seen, is one in the sense of being an individual self-contained
whole of experience. And its individuality means that it is the
systematic embodiment of a single coherent structure in a plurality of
elements or parts, which depend for their whole character upon the
fact that they are the embodiment of precisely this structure. If this is
so, we may say that degrees of reality mean the same thing as
degrees of individuality, and that a thing is real precisely to the same
extent to which it is truly individual.
A thing, that is, no matter of what kind, is really what it appears to
be, just in so far as the thing, as it appears for our knowledge, is
itself a self-contained and therefore unique systematic whole. Or, in
other words, just in so far as what we recognise as one thing shows
itself, in the face of philosophical criticism and analysis, to be a self-
contained systematic whole, so far are we truly apprehending that
thing as a manifestation of the fundamental character of Reality, of
seeing it as it really is, and so far does our knowledge give us
genuine Reality. On the other hand, just so far as what at first
seemed a self-contained whole is discovered by subsequent
analysis not to be so, so far have we failed to see the facts in their
true place in the single whole of Reality, and so far is our knowledge
affected with error and unreality. Or, again, the more truly anything is
a self-contained individual whole, the higher its place in the scale of
Reality.
When we ask what are the marks by which one thing may be
shown to be more of a true individual whole than another, we shall
find that they may be reduced to two, both of which we can easily
see to be in principle the same, though, owing to the limitations of
our insight, they do not always appear to coincide in a given case.
One thing is ceteris paribus more truly an individual whole than
another: (1) when the wealth of detailed content it embraces is
greater; (2) when the completeness of the unity with which it
embraces that detail is greater. Or, the degree of individuality
possessed by any system depends: (1) on its comprehensiveness;
(2) on its internal systematisation. The more a thing includes of
existence and the more harmoniously it includes it, the more
individual it is.
It is manifest, of course, that these two characteristics of a
systematic whole are mutually interdependent. For, precisely
because all Reality is ultimately a single coherent system, the more
there is outside any partial system the greater must be the
dependence of its constituents for their character upon their
connection with reality outside, and the less capable must the
system be of complete explanation from within itself. The more the
partial system embraces, the less will its constituents be determined
by relation to anything outside itself, and the more completely will its
organisation be explicable by reference to its own internal principle
of structure. That is, the greater the comprehensiveness of the
system, the completer in general will be its internal coherence. And,
conversely, the more completely the working of the whole system in
its details is explicable from within as the expression of a single
principle of internal structure, the less must be the dependence of its
contents on any external reality; and therefore, seeing that all reality
is ultimately interconnected, the less must be the extent of what lies
outside the system in question. That is, the greater the internal unity,
the greater in general the comprehensiveness of the system. Thus
ultimately the two criteria of individuality coincide.
§ 5. In practice, however, it constantly happens, as a consequence
of the fragmentary way in which our experiences come to us, that
comprehensiveness and thorough-going systematic unity seem to be
opposed to one another. Thus we can see, as a general principle,
that the systematic organisation of knowledge depends upon its
extent. The wider our knowledge, the greater on the whole the
degree to which it exhibits organic structure; the systematisation of
science and its extension ultimately go together. Yet at any one
moment in the development of knowledge the recognition of fresh
truths may necessitate a temporary introduction of disorganisation
and discrepancy among the accepted principles of science. Thus in
the history of geometry the recognised principles of the science were
temporarily disorganised by the admission of incommensurable
magnitudes which was forced upon the early Greek mathematicians
by the discovery that the side and diagonal of a square have no
common measure, and the discrepancy was only removed when it
became possible to revise the principles of the theory of numbers
itself. So again at the present day there is a real danger that
premature anxiety to give the study of Psychology precise
systematic character by an exact definition of its subject and its
relation to the various physical and mental sciences, may stand in
the way of the extension of our knowledge of the facts of psychical
life. We have often to purchase an important extension of knowledge
at the cost of temporary confusion of principles, and to be content to
wait for the future readjustment of facts to principles in the course of
subsequent progress.
So in our moral life we judge one man’s character more individual
than another’s, either on the ground of the superior breadth of his
interests, or of the superior consistency with which his interests are
wrought into a self-consistent whole. The man of many interests has
so far a truer individuality than the man of few, and again the man of
steady purpose than the man whose energies are dissipated in
seemingly conflicting pursuits. But the two criteria do not always, for
our insight, coincide. An increase in variety and breadth of interests
may be accompanied by a diminution in coherency of aim, and a
gain in coherency of aim appears often to be bought by
concentration upon a few special objects. And we should find it hard
or impossible to decide, where the two aspects of individuality
appear to fall thus apart, whether the man of many interests and
relatively dissipated energies, or the man of few interests and
intense concentration upon them, exhibits the higher individuality.
For what looked like self-dissipation in the pursuit of disconnected
objects might really be the systematic pursuit of a consistent purpose
too wide to be clearly apprehended in its unity either by
contemporary observers or by the actor himself, yet apparent
enough to the reflective historian reading the significance of a life by
its whole effect upon society, and what seemed at the time the single
object of the man of one idea might similarly be found in the light of
the sequel to be the hasty combination of radically inconsistent aims.
[69]

Such reflections, however, only show that our limited insight is

insufficient to assign to every appearance with certainty its own
place in the ordered system of appearances through which the single
Reality expresses itself. They do not touch our general position, that
where comprehensiveness and harmony can be seen to go together,
we are justified in using them as the measure of the individuality and
therefore of the reality of the partial system in which we discover
them. It is on such grounds, for instance, that we may safely
pronounce that an organism, which is the living unity of its members,
is more individual and therefore a higher reality that a mere
aggregate of pre-existing units, in which the nature of the parts is
wholly or mainly independent of the structure of the whole; and
again, that a mind consciously and systematically pursuing a
coherent self-chosen system of ends is more individual, and
therefore again a higher reality, than an organism reacting according
to the temporary character of its environment or its momentary
internal condition in ways which form no systematic execution of a
connected scheme of ends. And it is clear that, if only on this ground,
we should have to say that we are nearer the truth in thinking of the
individual whole of complete Reality as an organism than in thinking
of it as an aggregate, and nearer the truth still in thinking of it as a
mind. Similarly in our judgments upon our own lives and character.
So far as one life possesses more breadth and again more
conscious unity of aim than another, so far it is more truly individual,
and therefore a more adequate type of complete reality. Just so far
as I am individual, I am truly real. And just so far as I fall short of
systematic individuality, whether from the poverty of my interests or
their mutual incompatibility, the appearance of unity in my life is
illusory, and I must be pronounced an unreal appearance.
At this point we may observe our metaphysical criterion of reality
coincides with our ethical criterion of moral worth. For in morality too
we esteem one life worthier than another, either for the superior
comprehensiveness of its ideals or for the thoroughness with which
they are wrought into a harmonious whole of coherent purpose. The
better man is either the man of the wider ideal, or again the man of
completer and purer self-devotion to his ideal. And thus for Morality
the measure of our worth, as for Metaphysics the measure of our
reality, lies in our individuality. And for Morality no less than for
Metaphysics individuality is pre-eminently a thing of degrees. In both
cases, again, the same difficulty besets us as soon as we attempt to
use our criterion for application to particular cases. Its two aspects
fall apart; it is not always the more comprehensive ideal which is
served with the higher fidelity of purpose. And so our actual moral
judgments on the worth of particular men, like our metaphysical
judgments on the order of reality to which particular things belong,
are often necessarily uncertain and fluctuating. We rate one man
morally high for the comprehensive rationality of his ideals, though
he suffers from a lack of concentrated energy, another for the steady
and earnest purpose with which he follows what we perhaps deplore
as a contracted ideal.
§ 6. One more point of supreme importance concerning the
relation of the lesser individuals to the perfect individual which is the
absolute whole of Reality. Now that we have learned what is meant
by degrees of individuality, we can see that there can, in the last
resort, be only one perfect and complete individual, the whole of
Reality itself, and that the subordinate individuals can never be
wholly and entirely individual in themselves. For to be a complete
individual would be, as we have seen, to be a whole system
absolutely self-contained and explicable solely by reference to
internal structure. Whatever requires, for the full understanding of its
systematic character, reference to existence outside itself, we have
seen, must also, so long as it is considered apart from the rest of
existence, be internally wanting in complete systematic harmony,
and thus must fall doubly short of the ideal of individuality.
And precisely because the whole of experience is a single system,
no lesser system within the whole is entirely explicable in terms of its
own internal structure. For a full understanding of the nature of the
lesser system, and of the way in which it manifests a common
character through the variety of its elements, you have always, in the
last resort, to go outside the system itself, and take into account its
relation to the rest of the whole system of existence. And for that
very reason no subordinate individual, considered in itself, is a
completely coherent self-determined whole. For a limited knowledge
like our own, which has in the main to deal with subordinate systems
as we find them, and without that complete understanding of the
whole structure of Reality which would enable us to see their precise
place in the whole, the subordinate systems themselves, when
closely scanned by a resolute philosophical analysis, will inevitably
exhibit some degree of discrepancy and want of systematic unity.
Consider, for instance, such a system as is formed by the life-work
of a man of marked “individuality.” On the whole, the life of such a
man may fairly be said to be the systematic working out of a
consistent scheme of purposes. But this is, after all, only
approximately the truth. It is not the case that the nature of the
central or dominant purpose of the scheme is of itself enough to
determine the nature and order of the successive stages by which it
finds expression. We have to take into account factors in the man’s
“heredity,” and again in his social and physical environment which
form no part of the nature of his central dominant ideal and yet
influence the manner of its fulfilment. We are thus thrown back for
our full understanding of the “individual” system in question upon
circumstances which are, so far as that system is concerned,
“accidental,” i.e. which are equally with itself part of the whole
system of experienced fact, without our being able to see how it and
they form a wider coherent whole. The subordinate individual,
because incapable of explication solely from within, is in the end only
approximately “individual,” and we therefore fall into contradictions
whenever we isolate it from the rest of Reality and treat it as
absolutely individual and self-contained.
In dealing with subordinate wholes, we always, if we go far
enough, come to a point where we have to recognise their
dependence upon a realm of external fact which our knowledge fails
to see in its systematic relation with them, and has therefore to treat
as accidental or as an ultimate “collocation.” This is why, as has
already been said, full knowledge of our own aims and interests as a
genuine systematic whole would coincide with complete insight into
the structure of the whole universe. We may invert the sentiment of a
hackneyed verse, and say with equal truth that until you know what
God and man is, you cannot really know what the “flower in the
crannied wall” is. This is as much as to say that every appearance
must involve some element of contradiction for our philosophical
analysis precisely because we cannot in the end see fully how any
appearance is related to the whole of Reality. But we must carefully
remember that if appearances, taken by themselves, are
contradictory, this is not because they are appearances, but
because, as so taken, they are all to some extent mere appearance.
The conclusion of the whole matter is, that the individuality of
anything less than the ultimate whole of being is a matter of degree
and approximation. We shall be equally in error if we assume that
because no subordinate system is fully individual, some are not
more individual and therefore more real than others or if we declare
that, because whatever is real at all must be in its degree individual,
therefore every element of Reality is completely real in its isolation.
The first error is that of a one-sided Monism, the second that of an
equally one-sided Pluralism.
Once more we may note a point of coincidence between our
general metaphysical theory of individuality and our personal
experience as moral agents. In so far as each of us is truly an
individual, his aims and ends form a system with an internal unity
pervading its structure, and therefore capable of progressive
realisation as a system. Yet again, because each of us is less than
the whole of Reality, or, what is the same thing, because the
systematic unity of our inner life is never complete, and our totality of
interests relations, and aspirations never a completely self-contained
ordered whole, our ideals will always be found to contain aspects
which will not fully harmonise, elements which fall outside such a
unity of structure as it is possible to effect within the limit of our
single personality. And thus all our victories contain an inseparable
element of defeat. The defeated aspects of the self may no doubt,
and in general do, in proportion to the degree of our individuality,
belong to the “lower” and relatively more “untrue” self, yet they are
elements in the whole self, and their suppression is a genuine if
necessary self-suppression. There is a sense in which an aspect of
failure is an inevitable feature in the life of every subordinate and
therefore imperfect individual. Human life, even in the millennium, as
we rightly feel, would not be human life if the note of sadness were
altogether absent from it. Only in the single experience of the
absolute whole can the discordant notes be finally resolved into a
faultless harmony.
§ 7. Technically, we may mark the distinction between complete
and approximate individuality by saying that the absolute whole is an
infinite individual, whereas all lesser wholes are but finite individuals.
And here it is important to note carefully the true meaning of these
often much-abused terms. The infinite must not be confounded with
the indefinite or unfinished. Its fundamental property is not the
merely negative one of having no end or “last term,” but the positive
one of having an internal structure which is the harmonious and
complete expression of a single self-consistent principle. The finite,
again, is finite not primarily merely because it has a “last term,” i.e.
because there is something else outside it, but because its “last
term” is arbitrarily determined, i.e. determined by something other
than the principles of its internal structure. In other words, the
essential defect of the finite is that it is not solely determined by its
own structural principle.
 We can see this even in the simple case of the familiar “infinite
series” of arithmetic and algebra. Such a series as 1, 1/2, 1/4 ... is
“infinite” not merely because you never come to the last term, but
because its character is determined from within, solely by the
principle according to which each term is derived from the one
before it; that the series has no end is a simple consequence of this
positive property of self-determination. But suppose I take n terms of
this series and no more, where n is a specified number, the resulting
series is now finite, not primarily because there are more terms of
the same kind outside it, but because the number of terms to be
taken is not prescribed by the law of formation of the series, but fixed
with reference to some object independent of the principle of the
series itself. In other words, only the infinite is in the full sense of the
words a completely self-determined whole. The finite is the
imperfect, not primarily because there is something outside it, but
because its contents are not solely prescribed by the principle of
structure which they embody. I, for instance, am a finite being, not
principally or merely because there are other men in the world, but
because my ideas and purposes are not a fully coherent systematic
whole in themselves.[70]
The view we have taken of individuality and the distinction
between finite and infinite individuality is closely akin to some of the
most fundamental ideas in the philosophy of Leibnitz. It was the
doctrine of Leibnitz that each of his monads “represented” the nature
of the whole system of existence, i.e. repeated the structure of the
whole in its own special structure, from a particular “point of view.”
According to the fulness and clearness of the “representation,” i.e.
the adequacy with which the structure of the monad repeated the
structure of the whole system, the monads were classed as higher or
lower in the scale of existence. The clearer a monad’s representation
of the whole within itself, the greater the monad’s “activity”; the more
confused the representation, the greater its “passivity.” It followed
that, inasmuch as no created monad fully exhibits the systematic
structure of the whole of Reality within itself, every one contains
some element of “passivity,” and that to be “passive” primarily means
not to be affected by extraneous influences, but to contain internal
“confusion.”