起步 合作式多学科的框架流程式教学

“Getting Started” Teaching in a Collaborative Multidisciplinary Framework

撰文 （美国）卡尔 · 斯坦尼茨

Text by Carl STEINITZ(US)

翻译 邝志峰 林广思

Translation by KUANG Zhi-feng and LIN Guang-si

摘要：在卡尔·斯坦尼茨的景观规划和设计教学框架流程中，作为主要成果的“改变模型”的起步阶段是最艰难也是最重要的。在简述其教学框架流程的基本内容和使用方式之后，介绍了百慕大和墨西哥泰波兹特兰两个设计课题的教学过程，最后讨论了教学中团队合作的角色分工问题。

关键词：风景园林；风景园林教育；景观规划；情景规划

Abstract: In Carl Steinitz’s framework of Landscape Planning and Design teaching, the most difficult stage of “getting started” on the changes which will be proposed as the main “product” is considered as the most important stage. After the context and function introduction of his teaching framework, Carl presents the two teaching processes of the Bermuda and Tepotzotlan projects, and discusses the roles in collaboration during the teaching process.

Key words: Landscape Architecture; Landscape Architecture Education; Landscape Planning; Scenario Planning

1 概述

我在哈佛及其他高校从事有关跨学科合作、学期制的大尺度复杂景观规划和设计方面的设计课题和课程教学已有40多年。我也组织和教授过许多1-4天的设计工作坊。我已经在其他书籍和文献上介绍过我在安排自己大部分教学工作过程中的教学策略和框架流程。本文将着重阐述框架流程中一个相关主题，即作为主要成果的“改变模型”的起步阶段。这一阶段是最艰难也是最重要的阶段，因为假如开头不令人满意，结果也会如此。

我的教学方式通常要求学生参与团队合作，并且是参与多学科的大团队合作，其原因有许多，但主要源于工作坊或设计课题所关注的问题的范围和复杂度，以及许多个人任务需要协调配合。有时，团队会小至3个人，但有时也包含12到18个人的设计班，作为一个“完整团队”。

我工作和教学的框架流程（图01）是由我制订的，其中参考了我的几位同事的建议（1990， 2003， 2012）。框架流程包括6个提问，这6个提问在课程学习中会被多次问到。在设计某个领域上的学习研究方案时，答案即模型及其应用，对于不同的案例研究来说，各不相同。一些模型的方法可以是通用的，但模型的参数和数据是基于研究的时间地点的，探究模型的输出和选择的结果过程便是研究。

这并非一个线性的过程，而是会有数个重复的回路。然而，它以有组织的提问展开工作。这一过程通常分几个步骤进行，每个步骤会有不同的练习和应用方法，使得小组的学生共同推进工作。我认为，一个能够定位各项任务并使其互相关联的明晰结构，是重大合作项目不可或缺的。

这6个提问分别是：（1）如何从内容、空间和时间这3个方面来描述景观的状况？这个提问通过表述模型（representation models）回答，研究有赖于其中的数据。（2）景观是如何运作的？其中要素的功能与结构关系如何？这个提问由为研究内容中的数个分析过程提供信息的过程模型（process models）来回答。（3）当前景观是否状况良好？这个提问用评价模型（evaluation models）回答，这一模型取决于制定决策的参与方的文化认知。（4）景观可能如何变更——通过什么政策和行动，在何处与何时？这个提问用改变模型（change models）回答，其将会在研究中接受测试。它也是数据，是未来的假设。（5）改变可能带来什么不同？这个提问用影响模型（impact models）回答，它是过程模型在更改前提条件后输出的数据。（6）景观应如何改变？这个提问用决策模型（decision models）回答，像评价模型一样，有赖于制定决策的参与方的文化认知。

整个研究过程中，6个提问及其附带提问都会被提出3次：第一次用来明确工作的条件及范围（“为什么”的问题）；第二次用来确定研究的方法（ “怎么办”的问题）；第三次用来执行研究的方法（ “何物”、“何处”和“何时”的问题）。

第一次迭代①是为了理解研究的背景及范围。研究的过程从大范围的背景调研和关注的主要问题开始。6个提问的框架流程从上至下运用。区域的现状描述和表述得到检验，并获得对景观是如何运作的总体认识。同时，所关注的内容，即现行规划和政策干预及其潜在影响以及制定决策的程序和准则，都会被调查。

第二次迭代是确定研究的方法。在此阶段，研究框架流程的运用自下而上。设计“可选将来”的研究的系统方法会涉及到特别复杂、通常基于经验与判断的决策。理解公共和私人决策如何改变景观是制定研究方法的基础。政策制定者及其选民所能接受的影响的议题和标准需要被调查，可能影响未来变化的规划与政策选择的制定方式需要被识别，现存景观状况也必须被理解和考虑，结构与功能的景观过程需要研究；各模型是针对性的。一旦这些过程被理解，数据需要识别，即对数据的要求和适当的表述方式能够被识别。

第三次迭代是执行方法以及完成研究。在此阶段，框架流程在此自上而下运用。数据收集到一起，列表，便于研究。执行过程模型，并且评价现存景观，以作为评价改变的影响的底线，许多的“可选择的将来”被模拟，而它们的影响也需要评价。然后，决策者可以更好地了解到他们的选择在将来可能的影响。

决策是该区域从市民到最高级别的政府机构的利益相关方的责任。为了制定决策，必须提出问题并进行回答，拟定可选方案并进行审核。图02示意了调研团队和利益相关方之间的关系。研究是为了回应利益相关方提出的问题和选择。“可选择的将来”及其影响的评价结果会向利益相关方展示，以便他们评审。大量的决策过程必须先于任何主要的行动计划。

决策有两个极端：“同意”或“否决”。“否决”意味着框架流程要回归撤销，并要回到更早前的阶段。所有6个阶段均可以成为反馈的焦点；“更多的数据”、“更好的模型”、和“改变方案的再设计”都是多次应用的反馈战略。

偶然的“同意”（实际上也是一个“否决”）可能还导致研究尺度、规模或时间上转变。当尺度转变时，研究将会重新在框架流程的6个阶段上开展，但是一些模型会有所不同。过程继续运作，直到得到肯定的决定（即“同意”）。“同意”意味着实施，即“改变”进入新的表述模型阶段。

当在尺度和时间上重复和相互联系，框架流程将成为一个相对复杂的研究基础。不考虑复杂性，相同的提问将一次又一次出现。但是，模型、方法及答案在各自背景下都会有所变化。

虽然框架流程、问题及模型的设置显得井然有序，其在应用时并非常常如此。任何研究的路线都不可能是一马平川的。研究会有错误的开始，没有收获的结果，还有偶然的发现，但是，在结论得出之前，它一定经历过上述框架流程中的提问及模型。

框架流程并非只是理论。它有助于解决工作坊、设计课题或者应用式研究项目中的复杂设计问题。只有当其确实对使用者有用时方为之有用。这个框架流程已经在众多环境中多次应用，而也似乎是实用和健全的。

我们如何开始？

我经常告知我的学生们，并没有诸如“设计方法”或“规划方法”（我认为规划也是设计）这样的东西。相反，通常有许多方法，它们要根据第二次迭代中现存的议题和问题来进行挑选和改进。所有景观设计，无论规模和尺度，都有3组影响，需被考虑：场地历史和以前的设计方案，基地难以被改变的一些“现实条件”，以及应该被列入“可选择的方案”的“恒量”。

最初的场地调研在框架流程的首次迭代中是不可或缺的。它是由团队和个人负责且计划紧凑的工作过程。毫无疑问，调研时熟悉议题、地理和人口的任务是最重要的。学识渊博的人负责汇报，他们也常常因为观点不同而进行激烈辩论。每个晚上，整个团队都召开小会，常有高层次的咨询汇报或其他交流。我们从不试图对研究进行定性，这是场地调研过程中至关重要的。我主要告诉学生，我们进行场地调研，是为了观察和提问，而不是决定任何东西。我希望学生们能够比较正式地讨论研究的范围和任务。

2案例分析

2.1百慕大

图示法及其体系由我在1982年讲授的一个设计课题中演化而来。百慕大最近从大不列颠独立了出来。首任总理约翰·斯旺希望对这个小岛国的垃圾堆填场未来情况进行研究并将规划新的垃圾焚烧场，但是一座新的垃圾焚烧场从规划到投入使用将需要3年甚至更长的时间。垃圾堆填场的四周设有市政机构、大型湿地、为百慕大大部分地区提供食用水源的井场和重要的游乐场。它位于当地贫民居住区范围内。斯旺总理竞选时已经承诺要把垃圾场改为百慕大的中央公园。我请愿带领这个研究设计，提供有关场地能建设何种类型的公园和辅助设施的不同设想。提请得到接受，并且设计课题得到财政支持。学生们自愿参加课题，同时也意识它会部分以设计竞赛的形式组织且并非所有人的设计都能最后实现。

设计课题组前往百慕大考察了研究的地区（图03），为利益相关方进行了几次汇报和召开开放会议，期间提出的问题都进行了记录。有关项目的组成、环境设计和相关政策的想法都向学生展示。每天晚上，我都和学生碰面，让他们对提出的问题进行罗列和归类，并根据他们收到的提议或自己构思的方案制作成图示。这些图示都是在统一尺度下，用简单线条进行绘制。所有图示都要求匿名、共享，所有人都知道这些。

回到学校之后，在这个设计课题第一阶段，学生们对20项必须在设计中需要解决的议题达成了一致意见。这些议题分为两类：每个设计不可或缺的恒量，以及因图示方法变更而可能产生的变量。每对学生根据他们的选择而分配到不同的议题，并且要对每一项议题提供2-5个可供选择的方案策略。他们用黑色马克笔在透明纸上绘制了80来个图示，以便于进行挑选、叠加和查阅，并将这些图示戏称为“三明治”。

下一步是议题的选择和排序，这可通过德尔菲技术（Delphi technique）完成。图04展示了大表格中的小图示布局。恒量在最左边一列；变量则在首行，但它们以观察到的重要性排序，最重要者在左。选择性的图示位于表头下方，并根据同学们采用德尔菲方法的评估结果，依据成功可能性排序。那么，每个人都可根据以下方式解读各个图示的位置。每个恒量图示必须列入，最可能成功的设计策略将能从左至右、自上而下选择。如果意见协调，接下来可以有两件事：首先通过忽视次重要的议题来选择或者把它们降级到第二或第三选择，成为不是最重要的议题。这个过程在第三次设计课时上完成。

在设计课的下一阶段，要求各个学生在所选的合适图示上准备初步设计。学生们将进行一次抽签，而第一个胜出的学生可以在众多图示中挑选一个。下一名的学生要求与前面的学生选不一样的。因此，当这些14个被选中的图示重叠在照明台上时，将会有14个不同的初步设计图。这些图纸将在第二周周末的第4次课之后准备好。

在此阶段，要和设计课程的学生讨论教学道德上的特别话题。即便在最有组织或教师带领下的课题设计，学生依然有绝对的权利去发展自己的想法，并以自己的方式完成。这是一个应该公开讨论的话题，而学生必须明白，教师的目标是教授和测试能对学生有利和有用的方法。这个设计课题的重点不是鼓励学生的特殊的创新。我很清楚意识到一些同事和学生并不同意这种观点，但我对此坚定。教师的道德责任要求如此，要在师生之间的社会契约中公开讨论。

每个学生接着准备一个初步设计的实体模型。在第6周末，这些模型用标准化批量生产的材料以标准尺度展示，其展示风格由学生小组拟定。每个模型都可以拆解并放在纸皮箱中（图05）。

3名学生带着14个设计方案前往百慕大。他们向负责场地重建的人员汇报这些设计。经过慎重的考虑，百慕大委员会选定了其中3个设计进入下一阶段继续深化。

这个结果由前往百慕大的同学在一次汇报和讨论中向大家告知。方案没有入选的同学要加入入选方案小组，这3个入选方案将在学期末进行展示。3个方案团队的规模也基本一样，都是自主组织的。设计课程继续深化3个不同的设计策略（图06）。

在学期结束时，这3个设计在哈佛向斯旺总理以及其他百慕大的代表汇报。整个班级随后也被邀请返回百慕大汇报这3个最终设计。百慕大当时约为9万人口，而至少有近万人看见了其中一个汇报。斯旺总理及其委员会决定在特殊选举之前对3个方案进行投票。这并不意味着要将其中的一个学生方案进行施工建设，而是要确定公众对这几个方案中所体现的设计战略的偏好。投票顺利举行，C方案最终胜出，它最符合课题初期采用的图表法（图04）中左边的恒量以及较上行的最佳方案的要求。

2.2墨西哥泰波兹特兰

接下来的例子是墨西哥的泰波兹特兰，它是图示方法在不同尺度上以及不同成果的应用实例，但是它在基础手段上采用了一种数字技术。这是一个研究级的课题设计，由我与胡安·卡罗斯·瓦格斯-莫里诺(Juan Carlos Vargas-Moreno)于2004年~2005年在哈佛大学开设，合作者还有墨西哥自治都市大学(Universidad Autonoma Metropolitana ,简称UAM)的安尼波尔·菲戈洛瓦(Anibal Figueroa) 教授带领师生团队，并得到了当地市政府的全力配合。

泰波兹特兰是一个自治市，位于可能是世界上最大的城市（指墨西哥城——译注）的北部边界上。她位于通往墨西哥北部的主要高速路上，面临着巨大的发展压力。它的四周是大量的公共住房和发货仓库（并且包围了城市）。非正规住房正大量增建。整个墨西哥城的污水未经过处理，就通过灌溉渠或是地下破旧的管道系统流经泰波兹特兰。至今，自治市依然保持相对少数移民的特征，拥有农耕以及城市周边的大片国有林场。

泰波兹特兰城的主要景点是建于1584年的圣弗兰西斯·泽维尔教堂及修道院（St. Francis Xavier）。这个不同寻常的建筑综合体是国家殖民地艺术博物馆，也是主要的旅游胜地。墨西哥领导性环保非政府组织FUNDEA是该景点的资助者，它在自治市拥有大量土地储备。此外，该地区的部分区域被开发作为环保教育公园。泰波兹特兰已经被墨西哥旅游部评定为全国10佳旅游业发展地之一。

因此，有以下几个潜在的矛盾需要解决：住房、交通、用水、污水处理、旅游业、保育以及康乐，它们都处于快速变化、环境恶化、政治复杂和经济相对贫困的状况中。这些议题及项目涉及从自治市区域到细部的各种尺度，同时也需要扮演建设性的角色以帮助自治市塑造未来。

在为期5天的泰波兹特兰调研中，参与设计课题的同学们根据每日的会议、讨论、参观以及获得的信息，制定了一系列项目和政策。每个项目都有1个或多个学生提出，在每天调研结束前的头脑风暴会中汇报。但这些项目没有修改或被否决。当8天行程结束时，学生们已经确认了近200个项目。

为了考虑和进入系统，这些项目提案有一个特殊的协定。胡安·卡罗斯·瓦格斯-莫里诺设计了技术上的程序。这些项目首先被列入“项目清单”，通过EXCEL的电子工作表组建立数据，再手绘成大区域地图（尺寸为3m×6m），放置在设计课题工作间。该图是采用最新的高分辨率正射投影（orthophotography）以及多层透明塑料布（plastic sheets）打印的。正射投影能让学生定位和描述每个提议项目的地理情况，而塑料布允许在投影仪上以单独工作表的形式绘制项目草图。在EXCEL的电子工作表中，每个项目标注有编号、提议学生的名字以及能够确认项目是否拥有独特空间物理变化或政策的分类号。此外，每个项目纳入八色示意中分类中的一个或多个类别：与国家或自治市政府相关的、与社区相关的、交通、工业、生态（包括水文）、遗产、设施以及野生动物保护区。

在场地调研的最后一天，根据不同的类别，将学生分成若干小组，且要求他们要像专家一样在每一个类别最重要的项目中最多挑选出20个项目。这个限制确实让学生专注战略性和重点的议题。共有近80个项目入围下一轮开发。这些项目都利用ESRI ArcMap 9.0在GIS上进行图像数据化处理。每个项目的图像都在独立的图层上数据化，以其所属类别的颜色标记，而全部的属性色谱加载到EXCEL的电子工作表中。依靠这些独立的电子数据库，以及简易地在电子表格中挑选想要图层的序号，学生在利用ESRI’s ArcScene得到了三维视觉上的不同效果群组。正射影像的可视化效果，逐层覆盖在数字高程模型上，同时也被单独项目的图层以假设重要性的反顺序逐个覆盖。不同的效果群组，例如旅游或生态项目，都是首要探究的对象。这样能让学生们形象地看到研究区域中不同项目类型的累积效果。随后，学生们通过班级讨论，合并不同项目，得到3个方案情景，分别是依据旅游业、生态以及经济驱动的选择。每个情景都以三维可视化方式展示（图07-09），并且标识了一组项目的序号（例如项目：2，6，26，55，43）。这些可视化效果都会展示给当地的合作方以及政府代表，并且用于日后深化的讨论。这些工作在场地调研中都已完成。

随后，课题组还准备了更多更复杂的情景，进行互相比较，最终选定一个情景。这个情景会发展为一个可选的市政规划平面图（图10），而一些项目将会在细部尺度上深化（图11-12）。

3结语

设计课题的开始阶段主要依赖图示、德尔菲方法和判断，我已介绍了框架流程应用的两种变量。我在许多短期工作坊中采用过这种方法。在所有案例下，参与者都完全意识到，这个阶段是一个确认议题的探索过程，在很长时间内还需要更加彻底的框架流程方法，需要更好的数据以在不同设计尺度中进行项目应用。

很明显，在高度组织化的团队式设计课题结构中，尽管每个学生都参与了研究的每一个阶段，他们个人是不能也不可能完成所有的事情。因此，我们鼓励学生在班级会议上将他们所做的事情以及其他同学可能感兴趣的事情进行汇报。项目归属于整个小组的学生。荣誉也是属于大家的，按字母顺序将他们的名字进行排序。当然，我们希望项目中的某些环节可能让学生有机会说“这是我做的”。但这一情况很少出现，因为整个研究性设计都是集体的。

那么教师又如何？教师这个角色是多样而富有挑战的。很明显，有人是“监制人”——负责组织课题项目的开展。他需要个人兴趣以及责任心，但并不总能有成果。还有人是首席顾问，他为整个小组以及学生个人提供建议。这个角色也是必须的，但有时候学生也会求助于其他专家顾问，通常是其他教师。毫无疑问，还有一只重要的“看不见的手”，以保持观察，防止悲剧发生。学生们通常会过度雄心壮志。他们常常会低估一些不能预估但是经验判断会发生的问题的影响。此外，课题小组还需要重要的调解员角色，通常是有关社会问题以及组织事务。他们有明确的责任要确保每个人的个体教育需求在团队组织的范围之内得到满足。除了调解员，还有 “评论家”，但仅仅在学生复查和讨论进展中的工作成果之后才能进行点评。最后，还会有一个人负责勘误，确保项目不会失败并能在时间和金钱允许范围之内完成。然而，最艰难的角色是要故意放弃控制管理和设计上的决定，而让团队从经验学习。无论怎么说，学生都会从中收获到批判性的教学经验。

我不能说，这条路子及其诸多方法会总是有效而良好的。我清楚地知道它暗含着图纸泛滥以及判断的失误，以及其依赖于参与者对合作适应并且能快读画图和做判断。在最差的情况下，他们能提出问题而继续深入研究，收集数据以及得到可选的设计策略。然而，在我的经验中，最好的情况是，这些方法能导向清晰、高效和健全的“起步”。

注释：

①译注：每次按顺序完成一系列工作流程就叫做一次迭代。

作者简介：

卡尔·斯坦尼茨/ 哈佛大学设计研究生院风景园林与规划荣誉退休教授

译者简介：

邝志峰/1990年生/广东人/华南理工大学建筑学院本科生（广州 510641）

林广思/1977年生/广东人/华南理工大学建筑学院教师（广州 510641）

 

1 OVERVIEW

I have led and taught collaborative, multidisciplinary, semester-long studios on large and complex landscape planning and design problems for more than 40 years at Harvard, and sometimes also with other universities. I also have organized and taught many one-to-four day workshops. I have written about my teaching strategies and the framework within which I organize most of my work in other books and papers. In this paper I want to focus on a related theme in the framework, the most difficult stage of “getting started” on the changes which will be proposed as the main “product”. I consider this stage to be the most important of any project because if the beginning is unsatisfactory, then the ending must also be.

The reasons for my teaching in a manner which requires students to work in teams, and frequently in large multidisciplinary teams, are many but normally center upon the scope and complexity of the problem around which the workshop or studio is focused and the need for many individual tasks to be coordinated. Sometimes teams have been as small as three persons, and sometimes they have involved a studio class of 12 to 18 persons acting as “a team of the whole”.

The framework within which I work and teach was developed by me, with advice from several colleagues (1990, 2003, 2012), and is shown in Fig.01. The framework consists of six questions which are asked several times during the course of a study. In designing a study of an area, the answers-the models and their applications-are particular to the case study. Some modeling approaches can be general, but model parameters and data are local to the place and time of the study as are the issues and options whose consequences are being studied.

This is NOT a linear process, but one which has several iteration `loops”. It does, however, follow an organized sequence of questions. For several steps in this sequence there are different exercises and applied methods which cause the group of students to move forward together. I think that having a clear structure-a framework around which tasks can be identified and linked-is essential in a large and collaborative effort.

The six questions are:

1. How should the state of the landscape be described in content, space and time? This question is answered by representation models, the data upon which the study relies.

2. How does the landscape operate? What are the functional and structural relationships among its elements? This question is answered by process models which provide information for the several analyses which are the content of the study.

3. Is the current landscape working well? This question is answered by evaluation models, which are dependent upon cultural knowledge of the decision making stakeholders.

4. How might the landscape be altered-by what policies and actions, where and when? This question is answered by the change models which will be tested in this research. They are also data, as assumed for the future.

5. What difference might the changes cause? This question is answered by impact models, which are information produced by the process models under changed conditions.

6. How should the landscape be changed? This question is answered by decision models, which, like the evaluation models, are dependent upon the cultural knowledge of the decision making responsibility is theirs.

Over the course of the study, each of the six questions and its subsidiary questions are asked three times: first to define the context and scope of the work (the WHY? questions); second to identify the methods of study (the HOW? questions, and third, to implement the study method (the WHAT, WHERE and WHEN? questions).

The objective of the first iteration is to understand the context and scope of the study. The study process begins with a broad survey of the setting and major issues of concern. The six questions framework is used from top to bottom. Existing descriptions and representations of the region are examined and a general knowledge of how the landscape works is developed. Areas of concern, existing plans and policy interventions and their potential impacts, and decision making processes and criteria are investigated.

The aim of the second iteration is to define the methods of the study. In this stage, the framework is used from bottom to top. Designing the methodology for a study of alternative futures involves decisions that are especially complex, and which are most often based on experience and judgment. Basic to developing the methodology is an understanding of how public and private decisions to change the landscape are made. The issues and the criteria defining acceptable impacts that decision makers and their constituents apply are investigated. Ways of identifying planning and policy choices that may influence future change are identified. Existing landscape conditions must be understood and considered. Structural and functional landscape processes are studied and models are specified. Once the processes are understood, and data needs identified, requirements for data and appropriate means of representations can be identified.

The third iteration implements the methods and carries out the study. In this third stage, the framework is again used from top to bottom. Data are gathered and represented in a format useful for study purposes. Process models are implemented, and evaluate the existing landscape as a baseline from which to assess impacts of change, a number of alternative futures are simulated, and their impacts assessed. Decision makers can then better understand the likely future impacts of their choices.

Decision making is the responsibility of the region’s stakeholders, from the individual citizen to the highest levels of government. In order to make decisions, questions must be asked and answered, and options for choice must be framed and deliberated. Figure 2 shows the relationship between the research team and the stakeholders. The study is shaped to respond to the issues and choices posed by the stakeholders. The alternative futures and the results of the assessments of their impacts are presented for stakeholder review and the many decision processes which must precede any major action.

At the extreme, two decision choices present themselves: “no” and “yes.” A “no” implies a backward feedback loop in the framework and the need to alter a prior level. All six levels can be the focus of feedback; “more data,” “a better model” and “redesign of the proposed changes” are frequently applied feedback strategies.

A contingent “yes” decision (still a “no”) may also trigger a shift in the scale or size or timing of the study. In a scale shift, the study will again proceed through the six levels of the framework but the several types of model will be different. It will then continue until it achieves a positive (“yes”) decision. A “yes” decision implies implementation, and (one assumes) a forward-in-time change to new representation models.

When repeated and linked over scale and time, the framework may be the organizing basis of a very complex study. Regardless of complexity, the same questions are posed again and again. However, the models, their methods, and their answers vary according to the context in which they are used.

While the framework and its set of questions and models looks orderly and sequential, it is frequently not so in application. The line through any study is not a smooth path. It has false starts, dead ends, and serendipitous discoveries, but it does pass through the questions and models of the framework as described herein before decisions can be made.

A framework is not a theory. It can be a useful aid to the organization of a complex design problem in a workshop, studio or in an applied- research program. It is only as useful as it is seen as useful by the user(s). This framework has been adapted and used many times and in many contexts, and it seems to be useful and robust.

How does one start?

I teach my students that there is no such thing as “THE Design Method” or “THE Planning Method” (and I consider a plan to be a design). Rather, there are many methods and they must be chosen and adapted to issues and questions raised by the problem at hand in the second iteration of the framework. Every landscape design regardless of size or scale has three groups of influences which should be considered: the history of the place and past proposals, the “facts” of the area which are not likely to be changed, and the “constants” which should be incorporated into any proposed alternative.

An initial field trip is indispensible for the first “scoping” iteration of the framework. It is an intensively scheduled working period with both group and individual responsibilities. Without question the tasks associated with becoming familiar with issues, geography, and people are of prime importance. There are presentations by knowledgeable persons, and these are frequently in conflict with each other. The entire group meets every evening and there is a high level of debriefing and other communications. Of critical importance during the field trip is the absence of any collective attempt to define the study. I make a major point of telling the students that we are on the field trip to observe and ask questions, not to decide anything. I do not want the students to informally negotiate the scope and responsibilities of the project.

2 CASE STUDY

2.1 BERMUDA

The diagramming methods and their organization derive from a studio which I taught in 1982. Bermuda had recently achieved independence from Great Britain. The first Prime Minister, John Swan, requested a study of the future of the garbage dump of that small island nation. There was a plan to build a new waste incinerator but it would take three years (and, in reality, many more) for that project to become operational. The garbage dump was surrounded by civic institutions, a large wetland, the well fields which supplied drinking water to most of Bermuda, and important play fields. It was in the midst of the residential area of the poorest people in the country. A promise had been made as part Mr. Swan’s election campaign to transform the dump area into a central park for Bermuda. I offered to teach a studio which would illustrate different assumptions regarding what kind of park and ancillary facilities might be developed for the site, and this offer was accepted and the studio was financed. Students volunteered for the studio knowing that it would be organized with some aspects of a design competition and that not all of their individual designs would be carried forward to the end.

The studio traveled to Bermuda and visited the study area (Fig.03). There were several presentations and several open meetings for interested persons during which records were kept of the issues raised, and ideas for program elements, physical designs and policies were presented to the students. Each evening I met with the students and had them list and categorize the issues which had been raised, and also to prepare simple diagrams of every idea and proposal which they had been offered or which they themselves had. These diagrams were all simple line drawings to a standard scale. They were anonymous and were intended to be shared, and all students knew this.

Upon returning to the University, and in the first working session of the studio, the students agreed on a final list of about 20 issues which had to be resolved in any design. These were of two kinds: the constants which had to be incorporated into every design, and the variables, for which there might be alternative diagrammatic solutions. Pairs of students were assigned by their choice to the variable issues and were asked to produce between two and five alternative strategies regarding each issue. There were approximately 80 diagrams each drawn with permanent black marker on thin clear plastic so that they could easily be selected, overlain, and looked at together as a set or, as the students called them, “a sandwich”.

The next exercise was to rank order the issues and alternatives, and this was done using a Delphi technique. Figure 4 represents what was actually the laying out of the small diagrams on a very large table. The constants all are in the left most column. The variables are listed along the top row, but they are in the rank order of their perceived importance, with the most important being to the left. The alternatives diagrams are below their heading and are also in rank order of likely success as the judged by the class using Delphi methods. Thus one can interpret the positions of the diagrams on the table in the following way. Every constant diagram must be included and in addition, the most likely successful design strategy would be to select the top row of issue alternatives starting from the left. If compromises were to be made, they could be made the two ways: first, by choosing to ignore the less important issues, or, by dropping to second or third best alternatives for any issue but preferably not the more important ones. This process was completed at the end of the third studio class.

In the next phase of the studio, each individual student was required to prepare an initial design by selecting an appropriate set of the diagrams. A lottery was held and the number one winning student had first choice among the variables diagrams. Each subsequent student in the lottery ranking was required to choose a different set from all previous students. Thus, when the selected diagrams were overlain on a light table, there were 14 substantially different initial diagrammatic designs. These were available after the fourth studio class, at the end of the second week.

A special issue of pedagogic ethics had to be discussed with the studio students in this phase. Even in the most organized or faculty led studio, there is an absolute right for a student to explore his or her own ideas, and in his or her own way. This is an issue which must be discussed openly and the students must understand that an objective of the faculty member is the teaching and testing of a method which is expected to be of interest and use to the student. The priority in this studio is not the encouragement of the students’ idiosyncratic creativity. I am well aware that some colleagues and some students do not agree with this position but it is the one which I hold. The ethics of being a teacher require this to be openly stated, openly discussed, and somehow managed within the social contract between student and teacher.

Each student then prepared a physical model of his or her initial design. At the end of the sixth week, these were presented in a standard scale using standard and mass produced materials, and in a representational style organized by a student subcommittee. Each model could be segmented and placed in a shipping carton (Fig.05).

Fourteen designs were flown to Bermuda along with three students who presented these designs to the group of persons actually responsible for the redevelopment of the site. After careful consideration, the Bermudian committee decided that three of the designs should be moved forward to the next stage.

This was reported to all in a presentation and discussion by the students who went to Bermuda. The students whose designs were not chosen to go forward then had to join the team of one of the three designs which would be presented at the end of the semester. The teams were of approximately equal size and were organized on a volunteer basis, and the studio continued with three very different design strategies (Fig.06).

At the end of the semester, a presentation of these three designs was held at Harvard with Mr. Swan and other representatives of Bermuda present. The entire class was then invited to return to Bermuda to present the three final designs. Bermuda at that time had a population of approximately 90,000 persons and about 10,000 persons saw at least one of the several presentations made by the students. The committee and Mr. Swan then decided to place the choice of one of the three park concepts before the electorate in a special election. The intent was not to build one of the student designs but rather to identify the preferences of the general public for the strategies which were embedded in the design options. This election was held and it is interesting to note that the winning design C, was the one which most closely conformed to left hand section of constants and the upper row “best options” in the diagram layout with which the studio got started(Fig.04).

2.2 TEPOTZATLAN, MEXICO

The following example, Tepotzotlan Mexico, is an application of the diagramming method to a problem at different scales and with different products, but using a digital adaptation of the basic approach. This was a graduate-level studio which I taught in 2004-5 at Harvard with Juan Carlos Vargas-Moreno, and in collaboration with a faculty/student team from the Universidad Autonoma Metropolitana (UAM) in Mexico city led by Professor Anibal Figueroa, and with the full cooperation of the municipal government.

Tepotzotlan is a municipality at the northern edge of what may be the largest city in the world. It is facing enormous development pressures, as it is on the main highway to the north of Mexico. It has, and is surrounded by, considerable amounts of social housing and distribution warehouses. There are increasing amounts of `informal housing’. The untreated sewage of the entire Mexico City area flows via Tepotzotlan, some in canals and some in a broken pipe system under Tepotzotlan. Yet the municipality retains the character of a group of relatively small settlements, with some agriculture and large adjacent National Forest lands.

The main attraction of the town of Tepotzotlan is the church and monastery of St. Francis Xavier, founded in 1584. This extraordinary complex is now the national museum of colonial art, and a major Mexican tourist attraction. The sponsor, FUNDEA, is the leading Mexican environmentally-oriented NGO. It has a large landholding in the municipality, and this area is partly developed and operated as an environmental-education park. Tepotzotlan has been identified by the Mexican Ministry of Tourism as one of 10 national priority areas for tourism development.

Thus the several potential conflicts needing resolution: housing, transport, water, sewage treatment, tourism, conservation, and recreation, and all in a rapidly changing, environmentally degraded, politically complex and relatively poor economy. There were issues and “projects” ranging from metropolitan-regional to very detailed scales, and always the need to play a constructive part in helping the municipality to shape its future.

During the five-day visit to Tepotzotlan, Mexico, the students participating in the studio created a list of projects and policies as reactions to the daily meetings, discussion, visits and information that had been gathered. Each project was proposed by one or more students and presented in brainstorming sessions that were held at the end of each day. But they were not edited or rejected. By the end of the eight-day field visit, the students had identified around 200 projects.

The project proposals had a specific protocol in order to be considered and entered in the system. The technical process had been designed by Juan Carlos Vargas-Moreno. The projects were first entered in a “project list” composed in EXCEL spreadsheet and then diagrammed by hand on a large regional map of 3 by 6 meters size that was placed in the studio work space. The table-map was a large print of the most recent high-resolution orthophotography and several layers of transparent plastic sheet. While the orthophoto allowed the students to locate and describe the geography of each proposed project, the plastic sheets allowed the sketching of projects over the orthophoto in independent sheets. In the EXCEL spreadsheet, each project had a number, the name of student who proposed it, a classification that determined if the project was a spatially specific physical change or a policy. Furthermore each project had to be classified in one or more of eight color-coded categories: national or municipal government related, neighborhood related, transportation, industrial, ecological (including hydrology), heritage, utilities or wildlife restoration.

During the last day of the site visit, students were divided in groups corresponding to each category, and were asked to act as experts by selecting up to 20 of the most significant projects in each category. This limit certainly focused the students on the issues of strategy and priority. A new short-list of around eighty projects was selected for further development. These projects were then digitized as diagrams in a GIS employing ESRI ArcMap 9.0. Each project diagram was digitized as a separate layer in the color code of its assigned category and the full spectrum of attributes entered in the EXCEL spreadsheet. With this electronic data base of individual projects, and by simply selecting the number of wanted layers via the spreadsheet, the students created different clusters of projects as overlays in a 3-D visualization generated by ESRI’s ArcScene. The visualization featured the orthophotography draped over the digital elevation model and covered by the individual project layers in inverse order of presumed importance. Different clusters such as tourism or ecological-related projects were created as initial explorations. This allowed the students to visualize the cumulative effect of different projects and categories in the region of study. Later, through class discussion, three scenarios were developed by combining different projects. The three scenarios were: tourism, ecological and economically–driven alternatives. Each scenario was presented in a 3-D visualization (Fig.07) and coded as a group of project numbers (e. g. projects: 2, 6, 26, 55, and 43). These visualizations were presented to the local collaborators and government representatives, and discussed for future refinement. This had been accomplished within the site visit.

Later, several more scenarios of more complex objectives were prepared and compared before the studio team decided to focus on one. This was developed further into an alternative municipal plan (Fig.08), and several projects developed at much more detailed scales.

3 CONCLUSION

I have illustrated two variations of the application of the framework to a studio’s beginning phase which relies on a mix of diagramming and Delphi methods and of judgment. I have used these methods many times in short workshops. In all cases, the participants were made fully aware that this stage was intended as a scoping exploration with the objective of identifying the issues and needs for a more thorough use of the framework over a longer period of time, with better data and with application to projects at different scales of design.

It is obvious that one of the individual costs of this highly organized team-based studio structure is that each student cannot and does not do all tasks even though each student participates in each phase of the study. Because of this students are encouraged to make presentations to the class meetings of things which they are doing and which may be of interest to others. The project belongs to the students as a group. Credit is shared by the team in alphabetical order. There may be internal tasks in which certain pieces are broken down under individual authorship. Indeed it is expected that every project have some component in which each student can say, “I did this.” But these situations are known in advance because of the study design. Everything else is “we”.

And what about the faculty? The faculty roles are varied and challenging. Clearly there is the role of producer-of organizing the situation in which the studio project can occur. It takes a long time. It carries a certain personal interest and commitment and it is not always fully successful. There is a considerable consultant-in-chief role in which advice is sought both teams and individual students. It must be either given, or students directed to other expert consultants, frequently other faculty members. Without doubt, there is a substantial “hidden hand” role, in which constant observation is necessary to ensure against disaster. Students are often overly ambitious. They frequently underestimate the impacts of problems which are unforeseen but which experience knows may occur. There are also important mediation roles, frequently around social questions and issues of organization. There is the real responsibility of ensuring that the individual educational needs of individual students are met within the scope of the team organization. There is the faculty role as “critic”, but only after the students have reviewed and discussed the work in progress. Finally there is the legal responsibility of oversight, ensuring that the project is not a disaster and that it is completed within the constraints of time and money available. Yet the most difficult faculty role is that of purposefully abstaining from controlling the many difficult managerial and design decisions, and letting the group learn by experience. After all, it is these experiences which are among the critical educational lessons to be gained by the students.

I cannot claim that this approach and its several teaching methods will always work efficiently and well. I fully acknowledge that they are potentially open to diagrammatic exaggeration and errors of judgment, and that they are dependent upon the participants having sufficient comfort in working together and in making rapid diagrams and judgments. At worst, they can raise questions for further research, data development and alternative design strategies. However, in my experience and at best, they provide clear, rapid and robust ways of “getting started”.

Reference:

Bermuda, Department of Planning. The Pembroke Marsh Plan, 1987 Bermuda: Department of Planning, Government of Bermuda, 1987.

Steinitz, C. (1990) A Framework for Theory Applicable to the Education of Landscape Architects (and Other Environmental Design Professionals). Landscape Journal, 9, 136-143.

Steinitz, C et al (2003), Chapter 3, “The Framework for Alternative Futures Studies” , in Steinitz, Carl, H. Arias, S. Bassett, M. Flaxman, T. Goode, T. Maddock, D. Mouat, R. Peiser, and A. Shearer, Alternative Futures for Changing Landscapes: The Upper San Pedro River Basin in Arizona and Sonora, Island Press, Washington, D.C.

Steinitz, C., (2012), A Framework for Geodesign, Esri Press, 2012 (to be published in Chinese in 2013 by Peking University Press)

Steinitz, C., A. Figueroa, and G. Castorena, eds. <ital>Futuros Alternativos para Tepotzotlan/ Alternative Futures for Tepotzotlan.<ital> Mexico D.F., Mexico: Universidad Autonoma Metropolitana, 2010.

 

Biography:

Carl Steinitz is a Emeritus Professor of Landscape Architecture and Planning, Graduate School of Design,Harvard University.

 

About the Translators:

KUANG Zhi-feng, born in Guangdong Province in 1990, is a student of School of Architecture, South China University of Technology;

LIN Guang-si, born in 1977, Guangdong Province, is a teacher of School of Architecture, South China University of Technology.